Merge pull request #5674 from open-webui/dev

0.3.28

.dockerignore

@@ -10,7 +10,8 @@ node_modules
 vite.config.js.timestamp-*
 vite.config.ts.timestamp-*
 __pycache__
-.env
+.idea
+venv
 _old
 uploads
 .ipynb_checkpoints

.github/ISSUE_TEMPLATE/bug_report.md

@@ -8,36 +8,43 @@ assignees: ''
 
 # Bug Report
 
-## Description
+## Installation Method
 
-**Bug Summary:**
-[Provide a brief but clear summary of the bug]
-
-**Steps to Reproduce:**
-[Outline the steps to reproduce the bug. Be as detailed as possible.]
-
-**Expected Behavior:**
-[Describe what you expected to happen.]
-
-**Actual Behavior:**
-[Describe what actually happened.]
+[Describe the method you used to install the project, e.g., git clone, Docker, pip, etc.]
 
 ## Environment
 
-- **Open WebUI Version:** [e.g., 0.1.120]
-- **Ollama (if applicable):** [e.g., 0.1.30, 0.1.32-rc1]
+- **Open WebUI Version:** [e.g., v0.3.11]
+- **Ollama (if applicable):** [e.g., v0.2.0, v0.1.32-rc1]
 
 - **Operating System:** [e.g., Windows 10, macOS Big Sur, Ubuntu 20.04]
 - **Browser (if applicable):** [e.g., Chrome 100.0, Firefox 98.0]
 
-## Reproduction Details
-
 **Confirmation:**
 
 - [ ] I have read and followed all the instructions provided in the README.md.
 - [ ] I am on the latest version of both Open WebUI and Ollama.
 - [ ] I have included the browser console logs.
 - [ ] I have included the Docker container logs.
+- [ ] I have provided the exact steps to reproduce the bug in the "Steps to Reproduce" section below.
+
+## Expected Behavior:
+
+[Describe what you expected to happen.]
+
+## Actual Behavior:
+
+[Describe what actually happened.]
+
+## Description
+
+**Bug Summary:**
+[Provide a brief but clear summary of the bug]
+
+## Reproduction Details
+
+**Steps to Reproduce:**
+[Outline the steps to reproduce the bug. Be as detailed as possible.]
 
 ## Logs and Screenshots
 
@@ -47,13 +54,9 @@ assignees: ''
 **Docker Container Logs:**
 [Include relevant Docker container logs, if applicable]
 
-**Screenshots (if applicable):**
+**Screenshots/Screen Recordings (if applicable):**
 [Attach any relevant screenshots to help illustrate the issue]
 
-## Installation Method
-
-[Describe the method you used to install the project, e.g., manual installation, Docker, package manager, etc.]
-
 ## Additional Information
 
 [Include any additional details that may help in understanding and reproducing the issue. This could include specific configurations, error messages, or anything else relevant to the bug.]

.github/dependabot.yml

@@ -3,9 +3,10 @@ updates:
   - package-ecosystem: pip
     directory: '/backend'
     schedule:
-      interval: weekly
+      interval: monthly
+    target-branch: 'dev'
   - package-ecosystem: 'github-actions'
     directory: '/'
     schedule:
       # Check for updates to GitHub Actions every week
-      interval: 'weekly'
+      interval: monthly

.github/workflows/build-release.yml

@@ -10,61 +10,63 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-    - name: Checkout repository
-      uses: actions/checkout@v4
+      - name: Checkout repository
+        uses: actions/checkout@v4
 
-    - name: Check for changes in package.json
-      run: |
-        git diff --cached --diff-filter=d package.json || {
-          echo "No changes to package.json"
-          exit 1
-        }
-    
-    - name: Get version number from package.json
-      id: get_version
-      run: |
-        VERSION=$(jq -r '.version' package.json)
-        echo "::set-output name=version::$VERSION"
+      - name: Check for changes in package.json
+        run: |
+          git diff --cached --diff-filter=d package.json || {
+            echo "No changes to package.json"
+            exit 1
+          }
 
-    - name: Extract latest CHANGELOG entry
-      id: changelog
-      run: |
-        CHANGELOG_CONTENT=$(awk 'BEGIN {print_section=0;} /^## \[/ {if (print_section == 0) {print_section=1;} else {exit;}} print_section {print;}' CHANGELOG.md)
-        CHANGELOG_ESCAPED=$(echo "$CHANGELOG_CONTENT" | sed ':a;N;$!ba;s/\n/%0A/g')
-        echo "Extracted latest release notes from CHANGELOG.md:" 
-        echo -e "$CHANGELOG_CONTENT" 
-        echo "::set-output name=content::$CHANGELOG_ESCAPED"
+      - name: Get version number from package.json
+        id: get_version
+        run: |
+          VERSION=$(jq -r '.version' package.json)
+          echo "::set-output name=version::$VERSION"
 
-    - name: Create GitHub release
-      uses: actions/github-script@v7
-      with:
-        github-token: ${{ secrets.GITHUB_TOKEN }}
-        script: |
-          const changelog = `${{ steps.changelog.outputs.content }}`;
-          const release = await github.rest.repos.createRelease({
-            owner: context.repo.owner,
-            repo: context.repo.repo,
-            tag_name: `v${{ steps.get_version.outputs.version }}`,
-            name: `v${{ steps.get_version.outputs.version }}`,
-            body: changelog,
-          })
-          console.log(`Created release ${release.data.html_url}`)
+      - name: Extract latest CHANGELOG entry
+        id: changelog
+        run: |
+          CHANGELOG_CONTENT=$(awk 'BEGIN {print_section=0;} /^## \[/ {if (print_section == 0) {print_section=1;} else {exit;}} print_section {print;}' CHANGELOG.md)
+          CHANGELOG_ESCAPED=$(echo "$CHANGELOG_CONTENT" | sed ':a;N;$!ba;s/\n/%0A/g')
+          echo "Extracted latest release notes from CHANGELOG.md:" 
+          echo -e "$CHANGELOG_CONTENT" 
+          echo "::set-output name=content::$CHANGELOG_ESCAPED"
 
-    - name: Upload package to GitHub release
-      uses: actions/upload-artifact@v4
-      with:
-        name: package
-        path: .
-      env:
-        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      - name: Create GitHub release
+        uses: actions/github-script@v7
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          script: |
+            const changelog = `${{ steps.changelog.outputs.content }}`;
+            const release = await github.rest.repos.createRelease({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              tag_name: `v${{ steps.get_version.outputs.version }}`,
+              name: `v${{ steps.get_version.outputs.version }}`,
+              body: changelog,
+            })
+            console.log(`Created release ${release.data.html_url}`)
 
-    - name: Trigger Docker build workflow
-      uses: actions/github-script@v7
-      with:
-        script: |
-          github.rest.actions.createWorkflowDispatch({
-            owner: context.repo.owner,
-            repo: context.repo.repo,
-            workflow_id: 'docker-build.yaml',
-            ref: 'v${{ steps.get_version.outputs.version }}',
-          })
+      - name: Upload package to GitHub release
+        uses: actions/upload-artifact@v4
+        with:
+          name: package
+          path: |
+            .
+            !.git
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Trigger Docker build workflow
+        uses: actions/github-script@v7
+        with:
+          script: |
+            github.rest.actions.createWorkflowDispatch({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              workflow_id: 'docker-build.yaml',
+              ref: 'v${{ steps.get_version.outputs.version }}',
+            })

.github/workflows/docker-build.yaml

@@ -312,7 +312,7 @@ jobs:
 
   merge-main-images:
     runs-on: ubuntu-latest
-    needs: [ build-main-image ]
+    needs: [build-main-image]
     steps:
       # GitHub Packages requires the entire repository name to be in lowercase
       # although the repository owner has a lowercase username, this prevents some people from running actions after forking
@@ -364,10 +364,9 @@ jobs:
         run: |
           docker buildx imagetools inspect ${{ env.FULL_IMAGE_NAME }}:${{ steps.meta.outputs.version }}
 
-
   merge-cuda-images:
     runs-on: ubuntu-latest
-    needs: [ build-cuda-image ]
+    needs: [build-cuda-image]
     steps:
       # GitHub Packages requires the entire repository name to be in lowercase
       # although the repository owner has a lowercase username, this prevents some people from running actions after forking
@@ -423,7 +422,7 @@ jobs:
 
   merge-ollama-images:
     runs-on: ubuntu-latest
-    needs: [ build-ollama-image ]
+    needs: [build-ollama-image]
     steps:
       # GitHub Packages requires the entire repository name to be in lowercase
       # although the repository owner has a lowercase username, this prevents some people from running actions after forking

.github/workflows/integration-test.yml

@@ -15,6 +15,13 @@ jobs:
     name: Run Cypress Integration Tests
     runs-on: ubuntu-latest
     steps:
+      - name: Maximize build space
+        uses: AdityaGarg8/remove-unwanted-software@v4.1
+        with:
+          remove-android: 'true'
+          remove-haskell: 'true'
+          remove-codeql: 'true'
+
       - name: Checkout Repository
         uses: actions/checkout@v4
 
@@ -26,6 +33,10 @@ jobs:
             --file docker-compose.a1111-test.yaml \
             up --detach --build
 
+      - name: Delete Docker build cache
+        run: |
+          docker builder prune --all --force
+
       - name: Wait for Ollama to be up
         timeout-minutes: 5
         run: |
@@ -43,7 +54,7 @@ jobs:
         uses: cypress-io/github-action@v6
         with:
           browser: chrome
-          wait-on: "http://localhost:3000"
+          wait-on: 'http://localhost:3000'
           config: baseUrl=http://localhost:3000
 
       - uses: actions/upload-artifact@v4
@@ -67,6 +78,28 @@ jobs:
           path: compose-logs.txt
           if-no-files-found: ignore
 
+  # pytest:
+  #   name: Run Backend Tests
+  #   runs-on: ubuntu-latest
+  #   steps:
+  #     - uses: actions/checkout@v4
+
+  #     - name: Set up Python
+  #       uses: actions/setup-python@v4
+  #       with:
+  #         python-version: ${{ matrix.python-version }}
+
+  #     - name: Install dependencies
+  #       run: |
+  #         python -m pip install --upgrade pip
+  #         pip install -r backend/requirements.txt
+
+  #     - name: pytest run
+  #       run: |
+  #         ls -al
+  #         cd backend
+  #         PYTHONPATH=. pytest . -o log_cli=true -o log_cli_level=INFO
+
   migration_test:
     name: Run Migration Tests
     runs-on: ubuntu-latest
@@ -124,13 +157,13 @@ jobs:
           GLOBAL_LOG_LEVEL: debug
         run: |
           cd backend
-          uvicorn main:app --port "8080" --forwarded-allow-ips '*' &
+          uvicorn open_webui.main:app --port "8080" --forwarded-allow-ips '*' &
           UVICORN_PID=$!
-          # Wait up to 20 seconds for the server to start
-          for i in {1..20}; do
+          # Wait up to 40 seconds for the server to start
+          for i in {1..40}; do
               curl -s http://localhost:8080/api/config > /dev/null && break
               sleep 1
-              if [ $i -eq 20 ]; then
+              if [ $i -eq 40 ]; then
                   echo "Server failed to start"
                   kill -9 $UVICORN_PID
                   exit 1
@@ -151,7 +184,7 @@ jobs:
           DATABASE_URL: postgresql://postgres:postgres@localhost:5432/postgres
         run: |
           cd backend
-          uvicorn main:app --port "8081" --forwarded-allow-ips '*' &
+          uvicorn open_webui.main:app --port "8081" --forwarded-allow-ips '*' &
           UVICORN_PID=$!
           # Wait up to 20 seconds for the server to start
           for i in {1..20}; do
@@ -171,7 +204,7 @@ jobs:
           fi
 
           # Check that service will reconnect to postgres when connection will be closed
-          status_code=$(curl --write-out %{http_code} -s --output /dev/null http://localhost:8081/health)
+          status_code=$(curl --write-out %{http_code} -s --output /dev/null http://localhost:8081/health/db)
           if [[ "$status_code" -ne 200 ]] ; then
             echo "Server has failed before postgres reconnect check"
             exit 1
@@ -183,7 +216,7 @@ jobs:
             cur = conn.cursor(); \
             cur.execute('SELECT pg_terminate_backend(psa.pid) FROM pg_stat_activity psa WHERE datname = current_database() AND pid <> pg_backend_pid();')"
 
-          status_code=$(curl --write-out %{http_code} -s --output /dev/null http://localhost:8081/health)
+          status_code=$(curl --write-out %{http_code} -s --output /dev/null http://localhost:8081/health/db)
           if [[ "$status_code" -ne 200 ]] ; then
             echo "Server has not reconnected to postgres after connection was closed: returned status $status_code"
             exit 1
@@ -197,7 +230,7 @@ jobs:
 #          DATABASE_URL: mysql://root:mysql@localhost:3306/mysql
 #        run: |
 #          cd backend
-#          uvicorn main:app --port "8083" --forwarded-allow-ips '*' &
+#          uvicorn open_webui.main:app --port "8083" --forwarded-allow-ips '*' &
 #          UVICORN_PID=$!
 #          # Wait up to 20 seconds for the server to start
 #          for i in {1..20}; do

.github/workflows/release-pypi.yml

@@ -4,6 +4,7 @@ on:
   push:
     branches:
       - main # or whatever branch you want to use
+      - pypi-release
 
 jobs:
   release:

.gitignore

@@ -306,3 +306,4 @@ dist
 # cypress artifacts
 cypress/videos
 cypress/screenshots
+.vscode/settings.json

CHANGELOG.md

@@ -5,6 +5,429 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.3.28] - 2024-09-24
+
+### Fixed
+
+- **🔍 Web Search Functionality**: Corrected an issue where the web search option was not functioning properly.
+
+## [0.3.27] - 2024-09-24
+
+### Fixed
+
+- **🔄 Periodic Cleanup Error Resolved**: Fixed a critical RuntimeError related to the 'periodic_usage_pool_cleanup' coroutine, ensuring smooth and efficient performance post-pip install, correcting a persisting issue from version 0.3.26.
+- **📊 Enhanced LaTeX Rendering**: Improved rendering for LaTeX content, enhancing clarity and visual presentation in documents and mathematical models.
+
+## [0.3.26] - 2024-09-24
+
+### Fixed
+
+- **🔄 Event Loop Error Resolution**: Addressed a critical error where a missing running event loop caused 'periodic_usage_pool_cleanup' to fail with pip installs. This fix ensures smoother and more reliable updates and installations, enhancing overall system stability.
+
+## [0.3.25] - 2024-09-24
+
+### Fixed
+
+- **🖼️ Image Generation Functionality**: Resolved an issue where image generation was not functioning, restoring full capability for visual content creation.
+- **⚖️ Rate Response Corrections**: Addressed a problem where rate responses were not working, ensuring reliable feedback mechanisms are operational.
+
+## [0.3.24] - 2024-09-24
+
+### Added
+
+- **🚀 Rendering Optimization**: Significantly improved message rendering performance, enhancing user experience and webui responsiveness.
+- **💖 Favorite Response Feature in Chat Overview**: Users can now mark responses as favorite directly from the chat overview, enhancing ease of retrieval and organization of preferred responses.
+- **💬 Create Message Pairs with Shortcut**: Implemented creation of new message pairs using Cmd/Ctrl+Shift+Enter, making conversation editing faster and more intuitive.
+- **🌍 Expanded User Prompt Variables**: Added weekday, timezone, and language information variables to user prompts to match system prompt variables.
+- **🎵 Enhanced Audio Support**: Now includes support for 'audio/x-m4a' files, broadening compatibility with audio content within the platform.
+- **🔏 Model URL Search Parameter**: Added an ability to select a model directly via URL parameters, streamlining navigation and model access.
+- **📄 Enhanced PDF Citations**: PDF citations now open at the associated page, streamlining reference checks and document handling.
+- **🔧Use of Redis in Sockets**: Enhanced socket implementation to fully support Redis, enabling effective stateless instances suitable for scalable load balancing.
+- **🌍 Stream Individual Model Responses**: Allows specific models to have individualized streaming settings, enhancing performance and customization.
+- **🕒 Display Model Hash and Last Modified Timestamp for Ollama Models**: Provides critical model details directly in the Models workspace for enhanced tracking.
+- **❗ Update Info Notification for Admins**: Ensures administrators receive immediate updates upon login, keeping them informed of the latest changes and system statuses.
+
+### Fixed
+
+- **🗑️ Temporary File Handling On Windows**: Fixed an issue causing errors when accessing a temporary file being used by another process, Tools & Functions should now work as intended.
+- **🔓 Authentication Toggle Issue**: Resolved the malfunction where setting 'WEBUI_AUTH=False' did not appropriately disable authentication, ensuring that user experience and system security settings function as configured.
+- **🔧 Save As Copy Issue for Many Model Chats**: Resolved an error preventing users from save messages as copies in many model chats.
+- **🔒 Sidebar Closure on Mobile**: Resolved an issue where the mobile sidebar remained open after menu engagement, improving user interface responsivity and comfort.
+- **🛡️ Tooltip XSS Vulnerability**: Resolved a cross-site scripting (XSS) issue within tooltips, ensuring enhanced security and data integrity during user interactions.
+
+### Changed
+
+- **↩️ Deprecated Interface Stream Response Settings**: Moved to advanced parameters to streamline interface settings and enhance user clarity.
+- **⚙️ Renamed 'speedRate' to 'playbackRate'**: Standardizes terminology, improving usability and understanding in media settings.
+
+## [0.3.23] - 2024-09-21
+
+### Added
+
+- **🚀 WebSocket Redis Support**: Enhanced load balancing capabilities for multiple instance setups, promoting better performance and reliability in WebUI.
+- **🔧 Adjustable Chat Controls**: Introduced width-adjustable chat controls, enabling a personalized and more comfortable user interface.
+- **🌎 i18n Updates**: Improved and updated the Chinese translations.
+
+### Fixed
+
+- **🌐 Task Model Unloading Issue**: Modified task handling to use the Ollama /api/chat endpoint instead of OpenAI compatible endpoint, ensuring models stay loaded and ready with custom parameters, thus minimizing delays in task execution.
+- **📝 Title Generation Fix for OpenAI Compatible APIs**: Resolved an issue preventing the generation of titles, enhancing consistency and reliability when using multiple API providers.
+- **🗃️ RAG Duplicate Collection Issue**: Fixed a bug causing repeated processing of the same uploaded file. Now utilizes indexed files to prevent unnecessary duplications, optimizing resource usage.
+- **🖼️ Image Generation Enhancement**: Refactored OpenAI image generation endpoint to be asynchronous, preventing the WebUI from becoming unresponsive during processing, thus enhancing user experience.
+- **🔓 Downgrade Authlib**: Reverted Authlib to version 1.3.1 to address and resolve issues concerning OAuth functionality.
+
+### Changed
+
+- **🔍 Improved Message Interaction**: Enhanced the message node interface to allow for easier focus redirection with a simple click, streamlining user interaction.
+- **✨ Styling Refactor**: Updated WebUI styling for a cleaner, more modern look, enhancing user experience across the platform.
+
+## [0.3.22] - 2024-09-19
+
+### Added
+
+- **⭐ Chat Overview**: Introducing a node-based interactive messages diagram for improved visualization of conversation flows.
+- **🔗 Multiple Vector DB Support**: Now supports multiple vector databases, including the newly added Milvus support. Community contributions for additional database support are highly encouraged!
+- **📡 Experimental Non-Stream Chat Completion**: Experimental feature allowing the use of OpenAI o1 models, which do not support streaming, ensuring more versatile model deployment.
+- **🔍 Experimental Colbert-AI Reranker Integration**: Added support for "jinaai/jina-colbert-v2" as a reranker, enhancing search relevance and accuracy. Note: it may not function at all on low-spec computers.
+- **🕸️ ENABLE_WEBSOCKET_SUPPORT**: Added environment variable for instances to ignore websocket upgrades, stabilizing connections on platforms with websocket issues.
+- **🔊 Azure Speech Service Integration**: Added support for Azure Speech services for Text-to-Speech (TTS).
+- **🎚️ Customizable Playback Speed**: Playback speed control is now available in Call mode settings, allowing users to adjust audio playback speed to their preferences.
+- **🧠 Enhanced Error Messaging**: System now displays helpful error messages directly to users during chat completion issues.
+- **📂 Save Model as Transparent PNG**: Model profile images are now saved as PNGs, supporting transparency and improving visual integration.
+- **📱 iPhone Compatibility Adjustments**: Added padding to accommodate the iPhone navigation bar, improving UI display on these devices.
+- **🔗 Secure Response Headers**: Implemented security response headers, bolstering web application security.
+- **🔧 Enhanced AUTOMATIC1111 Settings**: Users can now configure 'CFG Scale', 'Sampler', and 'Scheduler' parameters directly in the admin settings, enhancing workflow flexibility without source code modifications.
+- **🌍 i18n Updates**: Enhanced translations for Chinese, Ukrainian, Russian, and French, fostering a better localized experience.
+
+### Fixed
+
+- **🛠️ Chat Message Deletion**: Resolved issues with chat message deletion, ensuring a smoother user interaction and system stability.
+- **🔢 Ordered List Numbering**: Fixed the incorrect ordering in lists.
+
+### Changed
+
+- **🎨 Transparent Icon Handling**: Allowed model icons to be displayed on transparent backgrounds, improving UI aesthetics.
+- **📝 Improved RAG Template**: Enhanced Retrieval-Augmented Generation template, optimizing context handling and error checking for more precise operation.
+
+## [0.3.21] - 2024-09-08
+
+### Added
+
+- **📊 Document Count Display**: Now displays the total number of documents directly within the dashboard.
+- **🚀 Ollama Embed API Endpoint**: Enabled /api/embed endpoint proxy support.
+
+### Fixed
+
+- **🐳 Docker Launch Issue**: Resolved the problem preventing Open-WebUI from launching correctly when using Docker.
+
+### Changed
+
+- **🔍 Enhanced Search Prompts**: Improved the search query generation prompts for better accuracy and user interaction, enhancing the overall search experience.
+
+## [0.3.20] - 2024-09-07
+
+### Added
+
+- **🌐 Translation Update**: Updated Catalan translations to improve user experience for Catalan speakers.
+
+### Fixed
+
+- **📄 PDF Download**: Resolved a configuration issue with fonts directory, ensuring PDFs are now downloaded with the correct formatting.
+- **🛠️ Installation of Tools & Functions Requirements**: Fixed a bug where necessary requirements for tools and functions were not properly installing.
+- **🔗 Inline Image Link Rendering**: Enabled rendering of images directly from links in chat.
+- **📞 Post-Call User Interface Cleanup**: Adjusted UI behavior to automatically close chat controls after a voice call ends, reducing screen clutter.
+- **🎙️ Microphone Deactivation Post-Call**: Addressed an issue where the microphone remained active after calls.
+- **✍️ Markdown Spacing Correction**: Corrected spacing in Markdown rendering, ensuring text appears neatly and as expected.
+- **🔄 Message Re-rendering**: Fixed an issue causing all response messages to re-render with each new message, now improving chat performance.
+
+### Changed
+
+- **🌐 Refined Web Search Integration**: Deprecated the Search Query Generation Prompt threshold; introduced a toggle button for "Enable Web Search Query Generation" allowing users to opt-in to using web search more judiciously.
+- **📝 Default Prompt Templates Update**: Emptied environment variable templates for search and title generation now default to the Open WebUI default prompt templates, simplifying configuration efforts.
+
+## [0.3.19] - 2024-09-05
+
+### Added
+
+- **🌐 Translation Update**: Improved Chinese translations.
+
+### Fixed
+
+- **📂 DATA_DIR Overriding**: Fixed an issue to avoid overriding DATA_DIR, preventing errors when directories are set identically, ensuring smoother operation and data management.
+- **🛠️ Frontmatter Extraction**: Fixed the extraction process for frontmatter in tools and functions.
+
+### Changed
+
+- **🎨 UI Styling**: Refined the user interface styling for enhanced visual coherence and user experience.
+
+## [0.3.18] - 2024-09-04
+
+### Added
+
+- **🛠️ Direct Database Execution for Tools & Functions**: Enhanced the execution of Python files for tools and functions, now directly loading from the database for a more streamlined backend process.
+
+### Fixed
+
+- **🔄 Automatic Rewrite of Import Statements in Tools & Functions**: Tool and function scripts that import 'utils', 'apps', 'main', 'config' will now automatically rename these with 'open_webui.', ensuring compatibility and consistency across different modules.
+- **🎨 Styling Adjustments**: Minor fixes in the visual styling to improve user experience and interface consistency.
+
+## [0.3.17] - 2024-09-04
+
+### Added
+
+- **🔄 Import/Export Configuration**: Users can now import and export webui configurations from admin settings > Database, simplifying setup replication across systems.
+- **🌍 Web Search via URL Parameter**: Added support for activating web search directly through URL by setting 'web-search=true'.
+- **🌐 SearchApi Integration**: Added support for SearchApi as an alternative web search provider, enhancing search capabilities within the platform.
+- **🔍 Literal Type Support in Tools**: Tools now support the Literal type.
+- **🌍 Updated Translations**: Improved translations for Chinese, Ukrainian, and Catalan.
+
+### Fixed
+
+- **🔧 Pip Install Issue**: Resolved the issue where pip install failed due to missing 'alembic.ini', ensuring smoother installation processes.
+- **🌃 Automatic Theme Update**: Fixed an issue where the color theme did not update dynamically with system changes.
+- **🛠️ User Agent in ComfyUI**: Added default headers in ComfyUI to fix access issues, improving reliability in network communications.
+- **🔄 Missing Chat Completion Response Headers**: Ensured proper return of proxied response headers during chat completion, improving API reliability.
+- **🔗 Websocket Connection Prioritization**: Modified socket.io configuration to prefer websockets and more reliably fallback to polling, enhancing connection stability.
+- **🎭 Accessibility Enhancements**: Added missing ARIA labels for buttons, improving accessibility for visually impaired users.
+- **⚖️ Advanced Parameter**: Fixed an issue ensuring that advanced parameters are correctly applied in all scenarios, ensuring consistent behavior of user-defined settings.
+
+### Changed
+
+- **🔁 Namespace Reorganization**: Reorganized all Python files under the 'open_webui' namespace to streamline the project structure and improve maintainability. Tools and functions importing from 'utils' should now use 'open_webui.utils'.
+- **🚧 Dependency Updates**: Updated several backend dependencies like 'aiohttp', 'authlib', 'duckduckgo-search', 'flask-cors', and 'langchain' to their latest versions, enhancing performance and security.
+
+## [0.3.16] - 2024-08-27
+
+### Added
+
+- **🚀 Config DB Migration**: Migrated configuration handling from config.json to the database, enabling high-availability setups and load balancing across multiple Open WebUI instances.
+- **🔗 Call Mode Activation via URL**: Added a 'call=true' URL search parameter enabling direct shortcuts to activate call mode, enhancing user interaction on mobile devices.
+- **✨ TTS Content Control**: Added functionality to control how message content is segmented for Text-to-Speech (TTS) generation requests, allowing for more flexible speech output options.
+- **😄 Show Knowledge Search Status**: Enhanced model usage transparency by displaying status when working with knowledge-augmented models, helping users understand the system's state during queries.
+- **👆 Click-to-Copy for Codespan**: Enhanced interactive experience in the WebUI by allowing users to click to copy content from code spans directly.
+- **🚫 API User Blocking via Model Filter**: Introduced the ability to block API users based on customized model filters, enhancing security and control over API access.
+- **🎬 Call Overlay Styling**: Adjusted call overlay styling on large screens to not cover the entire interface, but only the chat control area, for a more unobtrusive interaction experience.
+
+### Fixed
+
+- **🔧 LaTeX Rendering Issue**: Addressed an issue that affected the correct rendering of LaTeX.
+- **📁 File Leak Prevention**: Resolved the issue of uploaded files mistakenly being accessible across user chats.
+- **🔧 Pipe Functions with '**files**' Param**: Fixed issues with '**files**' parameter not functioning correctly in pipe functions.
+- **📝 Markdown Processing for RAG**: Fixed issues with processing Markdown in files.
+- **🚫 Duplicate System Prompts**: Fixed bugs causing system prompts to duplicate.
+
+### Changed
+
+- **🔋 Wakelock Permission**: Optimized the activation of wakelock to only engage during call mode, conserving device resources and improving battery performance during idle periods.
+- **🔍 Content-Type for Ollama Chats**: Added 'application/x-ndjson' content-type to '/api/chat' endpoint responses to match raw Ollama responses.
+- **✋ Disable Signups Conditionally**: Implemented conditional logic to disable sign-ups when 'ENABLE_LOGIN_FORM' is set to false.
+
+## [0.3.15] - 2024-08-21
+
+### Added
+
+- **🔗 Temporary Chat Activation**: Integrated a new URL parameter 'temporary-chat=true' to enable temporary chat sessions directly through the URL.
+- **🌄 ComfyUI Seed Node Support**: Introduced seed node support in ComfyUI for image generation, allowing users to specify node IDs for randomized seed assignment.
+
+### Fixed
+
+- **🛠️ Tools and Functions**: Resolved a critical issue where Tools and Functions were not properly functioning, restoring full capability and reliability to these essential features.
+- **🔘 Chat Action Button in Many Model Chat**: Fixed the malfunctioning of chat action buttons in many model chat environments, ensuring a smoother and more responsive user interaction.
+- **⏪ Many Model Chat Compatibility**: Restored backward compatibility for many model chats.
+
+## [0.3.14] - 2024-08-21
+
+### Added
+
+- **🛠️ Custom ComfyUI Workflow**: Deprecating several older environment variables, this enhancement introduces a new, customizable workflow for a more tailored user experience.
+- **🔀 Merge Responses in Many Model Chat**: Enhances the dialogue by merging responses from multiple models into a single, coherent reply, improving the interaction quality in many model chats.
+- **✅ Multiple Instances of Same Model in Chats**: Enhanced many model chat to support adding multiple instances of the same model.
+- **🔧 Quick Actions in Model Workspace**: Enhanced Shift key quick actions for hiding/unhiding and deleting models, facilitating a smoother workflow.
+- **🗨️ Markdown Rendering in User Messages**: User messages are now rendered in Markdown, enhancing readability and interaction.
+- **💬 Temporary Chat Feature**: Introduced a temporary chat feature, deprecating the old chat history setting to enhance user interaction flexibility.
+- **🖋️ User Message Editing**: Enhanced the user chat editing feature to allow saving changes without sending, providing more flexibility in message management.
+- **🛡️ Security Enhancements**: Various security improvements implemented across the platform to ensure safer user experiences.
+- **🌍 Updated Translations**: Enhanced translations for Chinese, Ukrainian, and Bahasa Malaysia, improving localization and user comprehension.
+
+### Fixed
+
+- **📑 Mermaid Rendering Issue**: Addressed issues with Mermaid chart rendering to ensure clean and clear visual data representation.
+- **🎭 PWA Icon Maskability**: Fixed the Progressive Web App icon to be maskable, ensuring proper display on various device home screens.
+- **🔀 Cloned Model Chat Freezing Issue**: Fixed a bug where cloning many model chats would cause freezing, enhancing stability and responsiveness.
+- **🔍 Generic Error Handling and Refinements**: Various minor fixes and refinements to address previously untracked issues, ensuring smoother operations.
+
+### Changed
+
+- **🖼️ Image Generation Refactor**: Overhauled image generation processes for improved efficiency and quality.
+- **🔨 Refactor Tool and Function Calling**: Refactored tool and function calling mechanisms for improved clarity and maintainability.
+- **🌐 Backend Library Updates**: Updated critical backend libraries including SQLAlchemy, uvicorn[standard], faster-whisper, bcrypt, and boto3 for enhanced performance and security.
+
+### Removed
+
+- **🚫 Deprecated ComfyUI Environment Variables**: Removed several outdated environment variables related to ComfyUI settings, simplifying configuration management.
+
+## [0.3.13] - 2024-08-14
+
+### Added
+
+- **🎨 Enhanced Markdown Rendering**: Significant improvements in rendering markdown, ensuring smooth and reliable display of LaTeX and Mermaid charts, enhancing user experience with more robust visual content.
+- **🔄 Auto-Install Tools & Functions Python Dependencies**: For 'Tools' and 'Functions', Open WebUI now automatically install extra python requirements specified in the frontmatter, streamlining setup processes and customization.
+- **🌀 OAuth Email Claim Customization**: Introduced an 'OAUTH_EMAIL_CLAIM' variable to allow customization of the default "email" claim within OAuth configurations, providing greater flexibility in authentication processes.
+- **📶 Websocket Reconnection**: Enhanced reliability with the capability to automatically reconnect when a websocket is closed, ensuring consistent and stable communication.
+- **🤳 Haptic Feedback on Support Devices**: Android devices now support haptic feedback for an immersive tactile experience during certain interactions.
+
+### Fixed
+
+- **🛠️ ComfyUI Performance Improvement**: Addressed an issue causing FastAPI to stall when ComfyUI image generation was active; now runs in a separate thread to prevent UI unresponsiveness.
+- **🔀 Session Handling**: Fixed an issue mandating session_id on client-side to ensure smoother session management and transitions.
+- **🖋️ Minor Bug Fixes and Format Corrections**: Various minor fixes including typo corrections, backend formatting improvements, and test amendments enhancing overall system stability and performance.
+
+### Changed
+
+- **🚀 Migration to SvelteKit 2**: Upgraded the underlying framework to SvelteKit version 2, offering enhanced speed, better code structure, and improved deployment capabilities.
+- **🧹 General Cleanup and Refactoring**: Performed broad cleanup and refactoring across the platform, improving code efficiency and maintaining high standards of code health.
+- **🚧 Integration Testing Improvements**: Modified how Cypress integration tests detect chat messages and updated sharing tests for better reliability and accuracy.
+- **📁 Standardized '.safetensors' File Extension**: Renamed the '.sft' file extension to '.safetensors' for ComfyUI workflows, standardizing file formats across the platform.
+
+### Removed
+
+- **🗑️ Deprecated Frontend Functions**: Removed frontend functions that were migrated to backend to declutter the codebase and reduce redundancy.
+
+## [0.3.12] - 2024-08-07
+
+### Added
+
+- **🔄 Sidebar Infinite Scroll**: Added an infinite scroll feature in the sidebar for more efficient chat navigation, reducing load times and enhancing user experience.
+- **🚀 Enhanced Markdown Rendering**: Support for rendering all code blocks and making images clickable for preview; codespan styling is also enhanced to improve readability and user interaction.
+- **🔒 Admin Shared Chat Visibility**: Admins no longer have default visibility over shared chats when ENABLE_ADMIN_CHAT_ACCESS is set to false, tightening security and privacy settings for users.
+- **🌍 Language Updates**: Added Malay (Bahasa Malaysia) translation and updated Catalan and Traditional Chinese translations to improve accessibility for more users.
+
+### Fixed
+
+- **📊 Markdown Rendering Issues**: Resolved issues with markdown rendering to ensure consistent and correct display across components.
+- **🛠️ Styling Issues**: Multiple fixes applied to styling throughout the application, improving the overall visual experience and interface consistency.
+- **🗃️ Modal Handling**: Fixed an issue where modals were not closing correctly in various model chat scenarios, enhancing usability and interface reliability.
+- **📄 Missing OpenAI Usage Information**: Resolved issues where usage statistics for OpenAI services were not being correctly displayed, ensuring users have access to crucial data for managing and monitoring their API consumption.
+- **🔧 Non-Streaming Support for Functions Plugin**: Fixed a functionality issue with the Functions plugin where non-streaming operations were not functioning as intended, restoring full capabilities for async and sync integration within the platform.
+- **🔄 Environment Variable Type Correction (COMFYUI_FLUX_FP8_CLIP)**: Corrected the data type of the 'COMFYUI_FLUX_FP8_CLIP' environment variable from string to boolean, ensuring environment settings apply correctly and enhance configuration management.
+
+### Changed
+
+- **🔧 Backend Dependency Updates**: Updated several backend dependencies such as boto3, pypdf, python-pptx, validators, and black, ensuring up-to-date security and performance optimizations.
+
+## [0.3.11] - 2024-08-02
+
+### Added
+
+- **📊 Model Information Display**: Added visuals for model selection, including images next to model names for more intuitive navigation.
+- **🗣 ElevenLabs Voice Adaptations**: Voice enhancements including support for ElevenLabs voice ID by name for personalized vocal interactions.
+- **⌨️ Arrow Keys Model Selection**: Users can now use arrow keys for quicker model selection, enhancing accessibility.
+- **🔍 Fuzzy Search in Model Selector**: Enhanced model selector with fuzzy search to locate models swiftly, including descriptions.
+- **🕹️ ComfyUI Flux Image Generation**: Added support for the new Flux image gen model; introduces environment controls like weight precision and CLIP model options in Settings.
+- **💾 Display File Size for Uploads**: Enhanced file interface now displays file size, preparing for upcoming upload restrictions.
+- **🎚️ Advanced Params "Min P"**: Added 'Min P' parameter in the advanced settings for customized model precision control.
+- **🔒 Enhanced OAuth**: Introduced custom redirect URI support for OAuth behind reverse proxies, enabling safer authentication processes.
+- **🖥 Enhanced Latex Rendering**: Adjustments made to latex rendering processes, now accurately detecting and presenting latex inputs from text.
+- **🌐 Internationalization**: Enhanced with new Romanian and updated Vietnamese and Ukrainian translations, helping broaden accessibility for international users.
+
+### Fixed
+
+- **🔧 Tags Handling in Document Upload**: Tags are now properly sent to the upload document handler, resolving issues with missing metadata.
+- **🖥️ Sensitive Input Fields**: Corrected browser misinterpretation of secure input fields, preventing misclassification as password fields.
+- **📂 Static Path Resolution in PDF Generation**: Fixed static paths that adjust dynamically to prevent issues across various environments.
+
+### Changed
+
+- **🎨 UI/UX Styling Enhancements**: Multiple minor styling updates for a cleaner and more intuitive user interface.
+- **🚧 Refactoring Various Components**: Numerous refactoring changes across styling, file handling, and function simplifications for clarity and performance.
+- **🎛️ User Valves Management**: Moved user valves from settings to direct chat controls for more user-friendly access during interactions.
+
+### Removed
+
+- **⚙️ Health Check Logging**: Removed verbose logging from the health checking processes to declutter logs and improve backend performance.
+
+## [0.3.10] - 2024-07-17
+
+### Fixed
+
+- **🔄 Improved File Upload**: Addressed the issue where file uploads lacked animation.
+- **💬 Chat Continuity**: Fixed a problem where existing chats were not functioning properly in some instances.
+- **🗂️ Chat File Reset**: Resolved the issue of chat files not resetting for new conversations, now ensuring a clean slate for each chat session.
+- **📁 Document Workspace Uploads**: Corrected the handling of document uploads in the workspace using the Files API.
+
+## [0.3.9] - 2024-07-17
+
+### Added
+
+- **📁 Files Chat Controls**: We've reverted to the old file handling behavior where uploaded files are always included. You can now manage files directly within the chat controls section, giving you the ability to remove files as needed.
+- **🔧 "Action" Function Support**: Introducing a new "Action" function to write custom buttons to the message toolbar. This feature enables more interactive messaging, with documentation coming soon.
+- **📜 Citations Handling**: For newly uploaded files in documents workspace, citations will now display the actual filename. Additionally, you can click on these filenames to open the file in a new tab for easier access.
+- **🛠️ Event Emitter and Call Updates**: Enhanced 'event_emitter' to allow message replacement and 'event_call' to support text input for Tools and Functions. Detailed documentation will be provided shortly.
+- **🎨 Styling Refactor**: Various styling updates for a cleaner and more cohesive user interface.
+- **🌐 Enhanced Translations**: Improved translations for Catalan, Ukrainian, and Brazilian Portuguese.
+
+### Fixed
+
+- **🔧 Chat Controls Priority**: Resolved an issue where Chat Controls values were being overridden by model information parameters. The priority is now Chat Controls, followed by Global Settings, then Model Settings.
+- **🪲 Debug Logs**: Fixed an issue where debug logs were not being logged properly.
+- **🔑 Automatic1111 Auth Key**: The auth key for Automatic1111 is no longer required.
+- **📝 Title Generation**: Ensured that the title generation runs only once, even when multiple models are in a chat.
+- **✅ Boolean Values in Params**: Added support for boolean values in parameters.
+- **🖼️ Files Overlay Styling**: Fixed the styling issue with the files overlay.
+
+### Changed
+
+- **⬆️ Dependency Updates**
+  - Upgraded 'pydantic' from version 2.7.1 to 2.8.2.
+  - Upgraded 'sqlalchemy' from version 2.0.30 to 2.0.31.
+  - Upgraded 'unstructured' from version 0.14.9 to 0.14.10.
+  - Upgraded 'chromadb' from version 0.5.3 to 0.5.4.
+
+## [0.3.8] - 2024-07-09
+
+### Added
+
+- **💬 Chat Controls**: Easily adjust parameters for each chat session, offering more precise control over your interactions.
+- **📌 Pinned Chats**: Support for pinned chats, allowing you to keep important conversations easily accessible.
+- **📄 Apache Tika Integration**: Added support for using Apache Tika as a document loader, enhancing document processing capabilities.
+- **🛠️ Custom Environment for OpenID Claims**: Allows setting custom claims for OpenID, providing more flexibility in user authentication.
+- **🔧 Enhanced Tools & Functions API**: Introduced 'event_emitter' and 'event_call', now you can also add citations for better documentation and tracking. Detailed documentation will be provided on our documentation website.
+- **↔️ Sideways Scrolling in Settings**: Settings tabs container now supports horizontal scrolling for easier navigation.
+- **🌑 Darker OLED Theme**: Includes a new, darker OLED theme and improved styling for the light theme, enhancing visual appeal.
+- **🌐 Language Updates**: Updated translations for Indonesian, German, French, and Catalan languages, expanding accessibility.
+
+### Fixed
+
+- **⏰ OpenAI Streaming Timeout**: Resolved issues with OpenAI streaming response using the 'AIOHTTP_CLIENT_TIMEOUT' setting, ensuring reliable performance.
+- **💡 User Valves**: Fixed malfunctioning user valves, ensuring proper functionality.
+- **🔄 Collapsible Components**: Addressed issues with collapsible components not working, restoring expected behavior.
+
+### Changed
+
+- **🗃️ Database Backend**: Switched from Peewee to SQLAlchemy for improved concurrency support, enhancing database performance.
+- **⬆️ ChromaDB Update**: Upgraded to version 0.5.3. Ensure your remote ChromaDB instance matches this version.
+- **🔤 Primary Font Styling**: Updated primary font to Archivo for better visual consistency.
+- **🔄 Font Change for Windows**: Replaced Arimo with Inter font for Windows users, improving readability.
+- **🚀 Lazy Loading**: Implemented lazy loading for 'faster_whisper' and 'sentence_transformers' to reduce startup memory usage.
+- **📋 Task Generation Payload**: Task generations now include only the "task" field in the body instead of "title".
+
+## [0.3.7] - 2024-06-29
+
+### Added
+
+- **🌐 Enhanced Internationalization (i18n)**: Newly introduced Indonesian translation, and updated translations for Turkish, Chinese, and Catalan languages to improve user accessibility.
+
+### Fixed
+
+- **🕵️‍♂️ Browser Language Detection**: Corrected the issue where the application was not properly detecting and adapting to the browser's language settings.
+- **🔐 OIDC Admin Role Assignment**: Fixed a bug where the admin role was not being assigned to the first user who signed up via OpenID Connect (OIDC).
+- **💬 Chat/Completions Endpoint**: Resolved an issue where the chat/completions endpoint was non-functional when the stream option was set to False.
+- **🚫 'WEBUI_AUTH' Configuration**: Addressed the problem where setting 'WEBUI_AUTH' to False was not being applied correctly.
+
+### Changed
+
+- **📦 Dependency Update**: Upgraded 'authlib' from version 1.3.0 to 1.3.1 to ensure better security and performance enhancements.
+
 ## [0.3.6] - 2024-06-27
 
 ### Added

Dockerfile

@@ -74,6 +74,10 @@ ENV RAG_EMBEDDING_MODEL="$USE_EMBEDDING_MODEL_DOCKER" \
 
 ## Hugging Face download cache ##
 ENV HF_HOME="/app/backend/data/cache/embedding/models"
+
+## Torch Extensions ##
+# ENV TORCH_EXTENSIONS_DIR="/.cache/torch_extensions"
+
 #### Other models ##########################################################
 
 WORKDIR /app/backend
@@ -96,7 +100,8 @@ RUN chown -R $UID:$GID /app $HOME
 RUN if [ "$USE_OLLAMA" = "true" ]; then \
     apt-get update && \
     # Install pandoc and netcat
-    apt-get install -y --no-install-recommends pandoc netcat-openbsd curl && \
+    apt-get install -y --no-install-recommends git build-essential pandoc netcat-openbsd curl && \
+    apt-get install -y --no-install-recommends gcc python3-dev && \
     # for RAG OCR
     apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 && \
     # install helper tools
@@ -107,8 +112,9 @@ RUN if [ "$USE_OLLAMA" = "true" ]; then \
     rm -rf /var/lib/apt/lists/*; \
     else \
     apt-get update && \
-    # Install pandoc and netcat
-    apt-get install -y --no-install-recommends pandoc netcat-openbsd curl jq && \
+    # Install pandoc, netcat and gcc
+    apt-get install -y --no-install-recommends git build-essential pandoc gcc netcat-openbsd curl jq && \
+    apt-get install -y --no-install-recommends gcc python3-dev && \
     # for RAG OCR
     apt-get install -y --no-install-recommends ffmpeg libsm6 libxext6 && \
     # cleanup
@@ -149,11 +155,12 @@ COPY --chown=$UID:$GID ./backend .
 
 EXPOSE 8080
 
-HEALTHCHECK CMD curl --silent --fail http://localhost:8080/health | jq -e '.status == true' || exit 1
+HEALTHCHECK CMD curl --silent --fail http://localhost:${PORT:-8080}/health | jq -ne 'input.status == true' || exit 1
 
 USER $UID:$GID
 
 ARG BUILD_HASH
 ENV WEBUI_BUILD_VERSION=${BUILD_HASH}
+ENV DOCKER true
 
 CMD [ "bash", "start.sh"]

README.md

@@ -37,7 +37,7 @@ Open WebUI is an [extensible](https://github.com/open-webui/pipelines), feature-
 
 - 📚 **Local RAG Integration**: Dive into the future of chat interactions with groundbreaking Retrieval Augmented Generation (RAG) support. This feature seamlessly integrates document interactions into your chat experience. You can load documents directly into the chat or add files to your document library, effortlessly accessing them using the `#` command before a query.
 
-- 🔍 **Web Search for RAG**: Perform web searches using providers like `SearXNG`, `Google PSE`, `Brave Search`, `serpstack`, `serper`, `Serply`, `DuckDuckGo` and `TavilySearch` and inject the results directly into your chat experience.
+- 🔍 **Web Search for RAG**: Perform web searches using providers like `SearXNG`, `Google PSE`, `Brave Search`, `serpstack`, `serper`, `Serply`, `DuckDuckGo`, `TavilySearch` and `SearchApi` and inject the results directly into your chat experience.
 
 - 🌐 **Web Browsing Capability**: Seamlessly integrate websites into your chat experience using the `#` command followed by a URL. This feature allows you to incorporate web content directly into your conversations, enhancing the richness and depth of your interactions.
 
@@ -59,11 +59,31 @@ Don't forget to explore our sibling project, [Open WebUI Community](https://open
 
 ## How to Install 🚀
 
-> [!NOTE]  
-> Please note that for certain Docker environments, additional configurations might be needed. If you encounter any connection issues, our detailed guide on [Open WebUI Documentation](https://docs.openwebui.com/) is ready to assist you.
+### Installation via Python pip 🐍
+
+Open WebUI can be installed using pip, the Python package installer. Before proceeding, ensure you're using **Python 3.11** to avoid compatibility issues.
+
+1. **Install Open WebUI**:
+   Open your terminal and run the following command to install Open WebUI:
+
+   ```bash
+   pip install open-webui
+   ```
+
+2. **Running Open WebUI**:
+   After installation, you can start Open WebUI by executing:
+
+   ```bash
+   open-webui serve
+   ```
+
+This will start the Open WebUI server, which you can access at [http://localhost:8080](http://localhost:8080)
 
 ### Quick Start with Docker 🐳
 
+> [!NOTE]  
+> Please note that for certain Docker environments, additional configurations might be needed. If you encounter any connection issues, our detailed guide on [Open WebUI Documentation](https://docs.openwebui.com/) is ready to assist you.
+
 > [!WARNING]
 > When using Docker to install Open WebUI, make sure to include the `-v open-webui:/app/backend/data` in your Docker command. This step is crucial as it ensures your database is properly mounted and prevents any loss of data.
 

backend/.gitignore

@@ -8,9 +8,5 @@ _test
 Pipfile
 !/data
 /data/*
-!/data/litellm
-/data/litellm/*
-!data/litellm/config.yaml
-
-!data/config.json
+/open_webui/data/*
 .webui_secret_key

backend/apps/audio/main.py

@@ -1,374 +0,0 @@
-import os
-import logging
-from fastapi import (
-    FastAPI,
-    Request,
-    Depends,
-    HTTPException,
-    status,
-    UploadFile,
-    File,
-    Form,
-)
-
-from fastapi.responses import StreamingResponse, JSONResponse, FileResponse
-
-from fastapi.middleware.cors import CORSMiddleware
-from faster_whisper import WhisperModel
-from pydantic import BaseModel
-
-import uuid
-import requests
-import hashlib
-from pathlib import Path
-import json
-
-from constants import ERROR_MESSAGES
-from utils.utils import (
-    decode_token,
-    get_current_user,
-    get_verified_user,
-    get_admin_user,
-)
-from utils.misc import calculate_sha256
-
-from config import (
-    SRC_LOG_LEVELS,
-    CACHE_DIR,
-    UPLOAD_DIR,
-    WHISPER_MODEL,
-    WHISPER_MODEL_DIR,
-    WHISPER_MODEL_AUTO_UPDATE,
-    DEVICE_TYPE,
-    AUDIO_STT_OPENAI_API_BASE_URL,
-    AUDIO_STT_OPENAI_API_KEY,
-    AUDIO_TTS_OPENAI_API_BASE_URL,
-    AUDIO_TTS_OPENAI_API_KEY,
-    AUDIO_STT_ENGINE,
-    AUDIO_STT_MODEL,
-    AUDIO_TTS_ENGINE,
-    AUDIO_TTS_MODEL,
-    AUDIO_TTS_VOICE,
-    AppConfig,
-)
-
-log = logging.getLogger(__name__)
-log.setLevel(SRC_LOG_LEVELS["AUDIO"])
-
-app = FastAPI()
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=["*"],
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-app.state.config = AppConfig()
-
-app.state.config.STT_OPENAI_API_BASE_URL = AUDIO_STT_OPENAI_API_BASE_URL
-app.state.config.STT_OPENAI_API_KEY = AUDIO_STT_OPENAI_API_KEY
-app.state.config.STT_ENGINE = AUDIO_STT_ENGINE
-app.state.config.STT_MODEL = AUDIO_STT_MODEL
-
-app.state.config.TTS_OPENAI_API_BASE_URL = AUDIO_TTS_OPENAI_API_BASE_URL
-app.state.config.TTS_OPENAI_API_KEY = AUDIO_TTS_OPENAI_API_KEY
-app.state.config.TTS_ENGINE = AUDIO_TTS_ENGINE
-app.state.config.TTS_MODEL = AUDIO_TTS_MODEL
-app.state.config.TTS_VOICE = AUDIO_TTS_VOICE
-
-# setting device type for whisper model
-whisper_device_type = DEVICE_TYPE if DEVICE_TYPE and DEVICE_TYPE == "cuda" else "cpu"
-log.info(f"whisper_device_type: {whisper_device_type}")
-
-SPEECH_CACHE_DIR = Path(CACHE_DIR).joinpath("./audio/speech/")
-SPEECH_CACHE_DIR.mkdir(parents=True, exist_ok=True)
-
-
-class TTSConfigForm(BaseModel):
-    OPENAI_API_BASE_URL: str
-    OPENAI_API_KEY: str
-    ENGINE: str
-    MODEL: str
-    VOICE: str
-
-
-class STTConfigForm(BaseModel):
-    OPENAI_API_BASE_URL: str
-    OPENAI_API_KEY: str
-    ENGINE: str
-    MODEL: str
-
-
-class AudioConfigUpdateForm(BaseModel):
-    tts: TTSConfigForm
-    stt: STTConfigForm
-
-
-from pydub import AudioSegment
-from pydub.utils import mediainfo
-
-
-def is_mp4_audio(file_path):
-    """Check if the given file is an MP4 audio file."""
-    if not os.path.isfile(file_path):
-        print(f"File not found: {file_path}")
-        return False
-
-    info = mediainfo(file_path)
-    if (
-        info.get("codec_name") == "aac"
-        and info.get("codec_type") == "audio"
-        and info.get("codec_tag_string") == "mp4a"
-    ):
-        return True
-    return False
-
-
-def convert_mp4_to_wav(file_path, output_path):
-    """Convert MP4 audio file to WAV format."""
-    audio = AudioSegment.from_file(file_path, format="mp4")
-    audio.export(output_path, format="wav")
-    print(f"Converted {file_path} to {output_path}")
-
-
-@app.get("/config")
-async def get_audio_config(user=Depends(get_admin_user)):
-    return {
-        "tts": {
-            "OPENAI_API_BASE_URL": app.state.config.TTS_OPENAI_API_BASE_URL,
-            "OPENAI_API_KEY": app.state.config.TTS_OPENAI_API_KEY,
-            "ENGINE": app.state.config.TTS_ENGINE,
-            "MODEL": app.state.config.TTS_MODEL,
-            "VOICE": app.state.config.TTS_VOICE,
-        },
-        "stt": {
-            "OPENAI_API_BASE_URL": app.state.config.STT_OPENAI_API_BASE_URL,
-            "OPENAI_API_KEY": app.state.config.STT_OPENAI_API_KEY,
-            "ENGINE": app.state.config.STT_ENGINE,
-            "MODEL": app.state.config.STT_MODEL,
-        },
-    }
-
-
-@app.post("/config/update")
-async def update_audio_config(
-    form_data: AudioConfigUpdateForm, user=Depends(get_admin_user)
-):
-    app.state.config.TTS_OPENAI_API_BASE_URL = form_data.tts.OPENAI_API_BASE_URL
-    app.state.config.TTS_OPENAI_API_KEY = form_data.tts.OPENAI_API_KEY
-    app.state.config.TTS_ENGINE = form_data.tts.ENGINE
-    app.state.config.TTS_MODEL = form_data.tts.MODEL
-    app.state.config.TTS_VOICE = form_data.tts.VOICE
-
-    app.state.config.STT_OPENAI_API_BASE_URL = form_data.stt.OPENAI_API_BASE_URL
-    app.state.config.STT_OPENAI_API_KEY = form_data.stt.OPENAI_API_KEY
-    app.state.config.STT_ENGINE = form_data.stt.ENGINE
-    app.state.config.STT_MODEL = form_data.stt.MODEL
-
-    return {
-        "tts": {
-            "OPENAI_API_BASE_URL": app.state.config.TTS_OPENAI_API_BASE_URL,
-            "OPENAI_API_KEY": app.state.config.TTS_OPENAI_API_KEY,
-            "ENGINE": app.state.config.TTS_ENGINE,
-            "MODEL": app.state.config.TTS_MODEL,
-            "VOICE": app.state.config.TTS_VOICE,
-        },
-        "stt": {
-            "OPENAI_API_BASE_URL": app.state.config.STT_OPENAI_API_BASE_URL,
-            "OPENAI_API_KEY": app.state.config.STT_OPENAI_API_KEY,
-            "ENGINE": app.state.config.STT_ENGINE,
-            "MODEL": app.state.config.STT_MODEL,
-        },
-    }
-
-
-@app.post("/speech")
-async def speech(request: Request, user=Depends(get_verified_user)):
-    body = await request.body()
-    name = hashlib.sha256(body).hexdigest()
-
-    file_path = SPEECH_CACHE_DIR.joinpath(f"{name}.mp3")
-    file_body_path = SPEECH_CACHE_DIR.joinpath(f"{name}.json")
-
-    # Check if the file already exists in the cache
-    if file_path.is_file():
-        return FileResponse(file_path)
-
-    headers = {}
-    headers["Authorization"] = f"Bearer {app.state.config.TTS_OPENAI_API_KEY}"
-    headers["Content-Type"] = "application/json"
-
-    try:
-        body = body.decode("utf-8")
-        body = json.loads(body)
-        body["model"] = app.state.config.TTS_MODEL
-        body = json.dumps(body).encode("utf-8")
-    except Exception as e:
-        pass
-
-    r = None
-    try:
-        r = requests.post(
-            url=f"{app.state.config.TTS_OPENAI_API_BASE_URL}/audio/speech",
-            data=body,
-            headers=headers,
-            stream=True,
-        )
-
-        r.raise_for_status()
-
-        # Save the streaming content to a file
-        with open(file_path, "wb") as f:
-            for chunk in r.iter_content(chunk_size=8192):
-                f.write(chunk)
-
-        with open(file_body_path, "w") as f:
-            json.dump(json.loads(body.decode("utf-8")), f)
-
-        # Return the saved file
-        return FileResponse(file_path)
-
-    except Exception as e:
-        log.exception(e)
-        error_detail = "Open WebUI: Server Connection Error"
-        if r is not None:
-            try:
-                res = r.json()
-                if "error" in res:
-                    error_detail = f"External: {res['error']['message']}"
-            except:
-                error_detail = f"External: {e}"
-
-        raise HTTPException(
-            status_code=r.status_code if r != None else 500,
-            detail=error_detail,
-        )
-
-
-@app.post("/transcriptions")
-def transcribe(
-    file: UploadFile = File(...),
-    user=Depends(get_current_user),
-):
-    log.info(f"file.content_type: {file.content_type}")
-
-    if file.content_type not in ["audio/mpeg", "audio/wav"]:
-        raise HTTPException(
-            status_code=status.HTTP_400_BAD_REQUEST,
-            detail=ERROR_MESSAGES.FILE_NOT_SUPPORTED,
-        )
-
-    try:
-        ext = file.filename.split(".")[-1]
-
-        id = uuid.uuid4()
-        filename = f"{id}.{ext}"
-
-        file_dir = f"{CACHE_DIR}/audio/transcriptions"
-        os.makedirs(file_dir, exist_ok=True)
-        file_path = f"{file_dir}/{filename}"
-
-        print(filename)
-
-        contents = file.file.read()
-        with open(file_path, "wb") as f:
-            f.write(contents)
-            f.close()
-
-        if app.state.config.STT_ENGINE == "":
-            whisper_kwargs = {
-                "model_size_or_path": WHISPER_MODEL,
-                "device": whisper_device_type,
-                "compute_type": "int8",
-                "download_root": WHISPER_MODEL_DIR,
-                "local_files_only": not WHISPER_MODEL_AUTO_UPDATE,
-            }
-
-            log.debug(f"whisper_kwargs: {whisper_kwargs}")
-
-            try:
-                model = WhisperModel(**whisper_kwargs)
-            except:
-                log.warning(
-                    "WhisperModel initialization failed, attempting download with local_files_only=False"
-                )
-                whisper_kwargs["local_files_only"] = False
-                model = WhisperModel(**whisper_kwargs)
-
-            segments, info = model.transcribe(file_path, beam_size=5)
-            log.info(
-                "Detected language '%s' with probability %f"
-                % (info.language, info.language_probability)
-            )
-
-            transcript = "".join([segment.text for segment in list(segments)])
-
-            data = {"text": transcript.strip()}
-
-            # save the transcript to a json file
-            transcript_file = f"{file_dir}/{id}.json"
-            with open(transcript_file, "w") as f:
-                json.dump(data, f)
-
-            print(data)
-
-            return data
-
-        elif app.state.config.STT_ENGINE == "openai":
-            if is_mp4_audio(file_path):
-                print("is_mp4_audio")
-                os.rename(file_path, file_path.replace(".wav", ".mp4"))
-                # Convert MP4 audio file to WAV format
-                convert_mp4_to_wav(file_path.replace(".wav", ".mp4"), file_path)
-
-            headers = {"Authorization": f"Bearer {app.state.config.STT_OPENAI_API_KEY}"}
-
-            files = {"file": (filename, open(file_path, "rb"))}
-            data = {"model": app.state.config.STT_MODEL}
-
-            print(files, data)
-
-            r = None
-            try:
-                r = requests.post(
-                    url=f"{app.state.config.STT_OPENAI_API_BASE_URL}/audio/transcriptions",
-                    headers=headers,
-                    files=files,
-                    data=data,
-                )
-
-                r.raise_for_status()
-
-                data = r.json()
-
-                # save the transcript to a json file
-                transcript_file = f"{file_dir}/{id}.json"
-                with open(transcript_file, "w") as f:
-                    json.dump(data, f)
-
-                print(data)
-                return data
-            except Exception as e:
-                log.exception(e)
-                error_detail = "Open WebUI: Server Connection Error"
-                if r is not None:
-                    try:
-                        res = r.json()
-                        if "error" in res:
-                            error_detail = f"External: {res['error']['message']}"
-                    except:
-                        error_detail = f"External: {e}"
-
-                raise HTTPException(
-                    status_code=r.status_code if r != None else 500,
-                    detail=error_detail,
-                )
-
-    except Exception as e:
-        log.exception(e)
-
-        raise HTTPException(
-            status_code=status.HTTP_400_BAD_REQUEST,
-            detail=ERROR_MESSAGES.DEFAULT(e),
-        )

backend/apps/images/utils/comfyui.py

@@ -1,250 +0,0 @@
-import websocket  # NOTE: websocket-client (https://github.com/websocket-client/websocket-client)
-import uuid
-import json
-import urllib.request
-import urllib.parse
-import random
-import logging
-
-from config import SRC_LOG_LEVELS
-
-log = logging.getLogger(__name__)
-log.setLevel(SRC_LOG_LEVELS["COMFYUI"])
-
-from pydantic import BaseModel
-
-from typing import Optional
-
-COMFYUI_DEFAULT_PROMPT = """
-{
-  "3": {
-    "inputs": {
-      "seed": 0,
-      "steps": 20,
-      "cfg": 8,
-      "sampler_name": "euler",
-      "scheduler": "normal",
-      "denoise": 1,
-      "model": [
-        "4",
-        0
-      ],
-      "positive": [
-        "6",
-        0
-      ],
-      "negative": [
-        "7",
-        0
-      ],
-      "latent_image": [
-        "5",
-        0
-      ]
-    },
-    "class_type": "KSampler",
-    "_meta": {
-      "title": "KSampler"
-    }
-  },
-  "4": {
-    "inputs": {
-      "ckpt_name": "model.safetensors"
-    },
-    "class_type": "CheckpointLoaderSimple",
-    "_meta": {
-      "title": "Load Checkpoint"
-    }
-  },
-  "5": {
-    "inputs": {
-      "width": 512,
-      "height": 512,
-      "batch_size": 1
-    },
-    "class_type": "EmptyLatentImage",
-    "_meta": {
-      "title": "Empty Latent Image"
-    }
-  },
-  "6": {
-    "inputs": {
-      "text": "Prompt",
-      "clip": [
-        "4",
-        1
-      ]
-    },
-    "class_type": "CLIPTextEncode",
-    "_meta": {
-      "title": "CLIP Text Encode (Prompt)"
-    }
-  },
-  "7": {
-    "inputs": {
-      "text": "Negative Prompt",
-      "clip": [
-        "4",
-        1
-      ]
-    },
-    "class_type": "CLIPTextEncode",
-    "_meta": {
-      "title": "CLIP Text Encode (Prompt)"
-    }
-  },
-  "8": {
-    "inputs": {
-      "samples": [
-        "3",
-        0
-      ],
-      "vae": [
-        "4",
-        2
-      ]
-    },
-    "class_type": "VAEDecode",
-    "_meta": {
-      "title": "VAE Decode"
-    }
-  },
-  "9": {
-    "inputs": {
-      "filename_prefix": "ComfyUI",
-      "images": [
-        "8",
-        0
-      ]
-    },
-    "class_type": "SaveImage",
-    "_meta": {
-      "title": "Save Image"
-    }
-  }
-}
-"""
-
-
-def queue_prompt(prompt, client_id, base_url):
-    log.info("queue_prompt")
-    p = {"prompt": prompt, "client_id": client_id}
-    data = json.dumps(p).encode("utf-8")
-    req = urllib.request.Request(f"{base_url}/prompt", data=data)
-    return json.loads(urllib.request.urlopen(req).read())
-
-
-def get_image(filename, subfolder, folder_type, base_url):
-    log.info("get_image")
-    data = {"filename": filename, "subfolder": subfolder, "type": folder_type}
-    url_values = urllib.parse.urlencode(data)
-    with urllib.request.urlopen(f"{base_url}/view?{url_values}") as response:
-        return response.read()
-
-
-def get_image_url(filename, subfolder, folder_type, base_url):
-    log.info("get_image")
-    data = {"filename": filename, "subfolder": subfolder, "type": folder_type}
-    url_values = urllib.parse.urlencode(data)
-    return f"{base_url}/view?{url_values}"
-
-
-def get_history(prompt_id, base_url):
-    log.info("get_history")
-    with urllib.request.urlopen(f"{base_url}/history/{prompt_id}") as response:
-        return json.loads(response.read())
-
-
-def get_images(ws, prompt, client_id, base_url):
-    prompt_id = queue_prompt(prompt, client_id, base_url)["prompt_id"]
-    output_images = []
-    while True:
-        out = ws.recv()
-        if isinstance(out, str):
-            message = json.loads(out)
-            if message["type"] == "executing":
-                data = message["data"]
-                if data["node"] is None and data["prompt_id"] == prompt_id:
-                    break  # Execution is done
-        else:
-            continue  # previews are binary data
-
-    history = get_history(prompt_id, base_url)[prompt_id]
-    for o in history["outputs"]:
-        for node_id in history["outputs"]:
-            node_output = history["outputs"][node_id]
-            if "images" in node_output:
-                for image in node_output["images"]:
-                    url = get_image_url(
-                        image["filename"], image["subfolder"], image["type"], base_url
-                    )
-                    output_images.append({"url": url})
-    return {"data": output_images}
-
-
-class ImageGenerationPayload(BaseModel):
-    prompt: str
-    negative_prompt: Optional[str] = ""
-    steps: Optional[int] = None
-    seed: Optional[int] = None
-    width: int
-    height: int
-    n: int = 1
-    cfg_scale: Optional[float] = None
-    sampler: Optional[str] = None
-    scheduler: Optional[str] = None
-    sd3: Optional[bool] = None
-
-
-def comfyui_generate_image(
-    model: str, payload: ImageGenerationPayload, client_id, base_url
-):
-    ws_url = base_url.replace("http://", "ws://").replace("https://", "wss://")
-
-    comfyui_prompt = json.loads(COMFYUI_DEFAULT_PROMPT)
-
-    if payload.cfg_scale:
-        comfyui_prompt["3"]["inputs"]["cfg"] = payload.cfg_scale
-
-    if payload.sampler:
-        comfyui_prompt["3"]["inputs"]["sampler"] = payload.sampler
-
-    if payload.scheduler:
-        comfyui_prompt["3"]["inputs"]["scheduler"] = payload.scheduler
-
-    if payload.sd3:
-        comfyui_prompt["5"]["class_type"] = "EmptySD3LatentImage"
-
-    comfyui_prompt["4"]["inputs"]["ckpt_name"] = model
-    comfyui_prompt["5"]["inputs"]["batch_size"] = payload.n
-    comfyui_prompt["5"]["inputs"]["width"] = payload.width
-    comfyui_prompt["5"]["inputs"]["height"] = payload.height
-
-    # set the text prompt for our positive CLIPTextEncode
-    comfyui_prompt["6"]["inputs"]["text"] = payload.prompt
-    comfyui_prompt["7"]["inputs"]["text"] = payload.negative_prompt
-
-    if payload.steps:
-        comfyui_prompt["3"]["inputs"]["steps"] = payload.steps
-
-    comfyui_prompt["3"]["inputs"]["seed"] = (
-        payload.seed if payload.seed else random.randint(0, 18446744073709551614)
-    )
-
-    try:
-        ws = websocket.WebSocket()
-        ws.connect(f"{ws_url}/ws?clientId={client_id}")
-        log.info("WebSocket connection established.")
-    except Exception as e:
-        log.exception(f"Failed to connect to WebSocket server: {e}")
-        return None
-
-    try:
-        images = get_images(ws, comfyui_prompt, client_id, base_url)
-    except Exception as e:
-        log.exception(f"Error while receiving images: {e}")
-        images = None
-
-    ws.close()
-
-    return images

backend/apps/socket/main.py

@@ -1,139 +0,0 @@
-import socketio
-import asyncio
-
-
-from apps.webui.models.users import Users
-from utils.utils import decode_token
-
-sio = socketio.AsyncServer(cors_allowed_origins=[], async_mode="asgi")
-app = socketio.ASGIApp(sio, socketio_path="/ws/socket.io")
-
-# Dictionary to maintain the user pool
-
-SESSION_POOL = {}
-USER_POOL = {}
-USAGE_POOL = {}
-# Timeout duration in seconds
-TIMEOUT_DURATION = 3
-
-
-@sio.event
-async def connect(sid, environ, auth):
-    user = None
-    if auth and "token" in auth:
-        data = decode_token(auth["token"])
-
-        if data is not None and "id" in data:
-            user = Users.get_user_by_id(data["id"])
-
-        if user:
-            SESSION_POOL[sid] = user.id
-            if user.id in USER_POOL:
-                USER_POOL[user.id].append(sid)
-            else:
-                USER_POOL[user.id] = [sid]
-
-            print(f"user {user.name}({user.id}) connected with session ID {sid}")
-
-            await sio.emit("user-count", {"count": len(set(USER_POOL))})
-            await sio.emit("usage", {"models": get_models_in_use()})
-
-
-@sio.on("user-join")
-async def user_join(sid, data):
-    print("user-join", sid, data)
-
-    auth = data["auth"] if "auth" in data else None
-
-    if auth and "token" in auth:
-        data = decode_token(auth["token"])
-
-        if data is not None and "id" in data:
-            user = Users.get_user_by_id(data["id"])
-
-        if user:
-
-            SESSION_POOL[sid] = user.id
-            if user.id in USER_POOL:
-                USER_POOL[user.id].append(sid)
-            else:
-                USER_POOL[user.id] = [sid]
-
-            print(f"user {user.name}({user.id}) connected with session ID {sid}")
-
-            await sio.emit("user-count", {"count": len(set(USER_POOL))})
-
-
-@sio.on("user-count")
-async def user_count(sid):
-    await sio.emit("user-count", {"count": len(set(USER_POOL))})
-
-
-def get_models_in_use():
-    # Aggregate all models in use
-    models_in_use = []
-    for model_id, data in USAGE_POOL.items():
-        models_in_use.append(model_id)
-
-    return models_in_use
-
-
-@sio.on("usage")
-async def usage(sid, data):
-
-    model_id = data["model"]
-
-    # Cancel previous callback if there is one
-    if model_id in USAGE_POOL:
-        USAGE_POOL[model_id]["callback"].cancel()
-
-    # Store the new usage data and task
-
-    if model_id in USAGE_POOL:
-        USAGE_POOL[model_id]["sids"].append(sid)
-        USAGE_POOL[model_id]["sids"] = list(set(USAGE_POOL[model_id]["sids"]))
-
-    else:
-        USAGE_POOL[model_id] = {"sids": [sid]}
-
-    # Schedule a task to remove the usage data after TIMEOUT_DURATION
-    USAGE_POOL[model_id]["callback"] = asyncio.create_task(
-        remove_after_timeout(sid, model_id)
-    )
-
-    # Broadcast the usage data to all clients
-    await sio.emit("usage", {"models": get_models_in_use()})
-
-
-async def remove_after_timeout(sid, model_id):
-    try:
-        await asyncio.sleep(TIMEOUT_DURATION)
-        if model_id in USAGE_POOL:
-            print(USAGE_POOL[model_id]["sids"])
-            USAGE_POOL[model_id]["sids"].remove(sid)
-            USAGE_POOL[model_id]["sids"] = list(set(USAGE_POOL[model_id]["sids"]))
-
-            if len(USAGE_POOL[model_id]["sids"]) == 0:
-                del USAGE_POOL[model_id]
-
-            # Broadcast the usage data to all clients
-            await sio.emit("usage", {"models": get_models_in_use()})
-    except asyncio.CancelledError:
-        # Task was cancelled due to new 'usage' event
-        pass
-
-
-@sio.event
-async def disconnect(sid):
-    if sid in SESSION_POOL:
-        user_id = SESSION_POOL[sid]
-        del SESSION_POOL[sid]
-
-        USER_POOL[user_id].remove(sid)
-
-        if len(USER_POOL[user_id]) == 0:
-            del USER_POOL[user_id]
-
-        await sio.emit("user-count", {"count": len(USER_POOL)})
-    else:
-        print(f"Unknown session ID {sid} disconnected")

backend/apps/webui/internal/db.py

@@ -1,54 +0,0 @@
-import os
-import logging
-import json
-
-from peewee import *
-from peewee_migrate import Router
-
-from apps.webui.internal.wrappers import register_connection
-from config import SRC_LOG_LEVELS, DATA_DIR, DATABASE_URL, BACKEND_DIR
-
-log = logging.getLogger(__name__)
-log.setLevel(SRC_LOG_LEVELS["DB"])
-
-
-class JSONField(TextField):
-    def db_value(self, value):
-        return json.dumps(value)
-
-    def python_value(self, value):
-        if value is not None:
-            return json.loads(value)
-
-
-# Check if the file exists
-if os.path.exists(f"{DATA_DIR}/ollama.db"):
-    # Rename the file
-    os.rename(f"{DATA_DIR}/ollama.db", f"{DATA_DIR}/webui.db")
-    log.info("Database migrated from Ollama-WebUI successfully.")
-else:
-    pass
-
-
-# The `register_connection` function encapsulates the logic for setting up
-# the database connection based on the connection string, while `connect`
-# is a Peewee-specific method to manage the connection state and avoid errors
-# when a connection is already open.
-try:
-    DB = register_connection(DATABASE_URL)
-    log.info(f"Connected to a {DB.__class__.__name__} database.")
-except Exception as e:
-    log.error(f"Failed to initialize the database connection: {e}")
-    raise
-
-router = Router(
-    DB,
-    migrate_dir=BACKEND_DIR / "apps" / "webui" / "internal" / "migrations",
-    logger=log,
-)
-router.run()
-try:
-    DB.connect(reuse_if_open=True)
-except OperationalError as e:
-    log.info(f"Failed to connect to database again due to: {e}")
-    pass

backend/apps/webui/internal/migrations/README.md

@@ -1,21 +0,0 @@
-# Database Migrations
-
-This directory contains all the database migrations for the web app.
-Migrations are done using the [`peewee-migrate`](https://github.com/klen/peewee_migrate) library.
-
-Migrations are automatically ran at app startup.
-
-## Creating a migration
-
-Have you made a change to the schema of an existing model?
-You will need to create a migration file to ensure that existing databases are updated for backwards compatibility.
-
-1. Have a database file (`webui.db`) that has the old schema prior to any of your changes.
-2. Make your changes to the models.
-3. From the `backend` directory, run the following command:
-   ```bash
-   pw_migrate create --auto --auto-source apps.webui.models --database sqlite:///${SQLITE_DB} --directory apps/web/internal/migrations ${MIGRATION_NAME}
-   ```
-   - `$SQLITE_DB` should be the path to the database file.
-   - `$MIGRATION_NAME` should be a descriptive name for the migration.
-4. The migration file will be created in the `apps/web/internal/migrations` directory.

backend/apps/webui/main.py

@@ -1,337 +0,0 @@
-from fastapi import FastAPI, Depends
-from fastapi.routing import APIRoute
-from fastapi.responses import StreamingResponse
-from fastapi.middleware.cors import CORSMiddleware
-from starlette.middleware.sessions import SessionMiddleware
-
-from apps.webui.routers import (
-    auths,
-    users,
-    chats,
-    documents,
-    tools,
-    models,
-    prompts,
-    configs,
-    memories,
-    utils,
-    files,
-    functions,
-)
-from apps.webui.models.functions import Functions
-from apps.webui.utils import load_function_module_by_id
-from utils.misc import stream_message_template
-
-from config import (
-    WEBUI_BUILD_HASH,
-    SHOW_ADMIN_DETAILS,
-    ADMIN_EMAIL,
-    WEBUI_AUTH,
-    DEFAULT_MODELS,
-    DEFAULT_PROMPT_SUGGESTIONS,
-    DEFAULT_USER_ROLE,
-    ENABLE_SIGNUP,
-    USER_PERMISSIONS,
-    WEBHOOK_URL,
-    WEBUI_AUTH_TRUSTED_EMAIL_HEADER,
-    WEBUI_AUTH_TRUSTED_NAME_HEADER,
-    JWT_EXPIRES_IN,
-    WEBUI_BANNERS,
-    ENABLE_COMMUNITY_SHARING,
-    AppConfig,
-)
-
-import inspect
-import uuid
-import time
-import json
-
-from typing import Iterator, Generator
-from pydantic import BaseModel
-
-app = FastAPI()
-
-origins = ["*"]
-
-app.state.config = AppConfig()
-
-app.state.config.ENABLE_SIGNUP = ENABLE_SIGNUP
-app.state.config.JWT_EXPIRES_IN = JWT_EXPIRES_IN
-app.state.AUTH_TRUSTED_EMAIL_HEADER = WEBUI_AUTH_TRUSTED_EMAIL_HEADER
-app.state.AUTH_TRUSTED_NAME_HEADER = WEBUI_AUTH_TRUSTED_NAME_HEADER
-
-
-app.state.config.SHOW_ADMIN_DETAILS = SHOW_ADMIN_DETAILS
-app.state.config.ADMIN_EMAIL = ADMIN_EMAIL
-
-
-app.state.config.DEFAULT_MODELS = DEFAULT_MODELS
-app.state.config.DEFAULT_PROMPT_SUGGESTIONS = DEFAULT_PROMPT_SUGGESTIONS
-app.state.config.DEFAULT_USER_ROLE = DEFAULT_USER_ROLE
-app.state.config.USER_PERMISSIONS = USER_PERMISSIONS
-app.state.config.WEBHOOK_URL = WEBHOOK_URL
-app.state.config.BANNERS = WEBUI_BANNERS
-
-app.state.config.ENABLE_COMMUNITY_SHARING = ENABLE_COMMUNITY_SHARING
-
-app.state.MODELS = {}
-app.state.TOOLS = {}
-app.state.FUNCTIONS = {}
-
-app.add_middleware(
-    CORSMiddleware,
-    allow_origins=origins,
-    allow_credentials=True,
-    allow_methods=["*"],
-    allow_headers=["*"],
-)
-
-
-app.include_router(configs.router, prefix="/configs", tags=["configs"])
-app.include_router(auths.router, prefix="/auths", tags=["auths"])
-app.include_router(users.router, prefix="/users", tags=["users"])
-app.include_router(chats.router, prefix="/chats", tags=["chats"])
-
-app.include_router(documents.router, prefix="/documents", tags=["documents"])
-app.include_router(models.router, prefix="/models", tags=["models"])
-app.include_router(prompts.router, prefix="/prompts", tags=["prompts"])
-
-app.include_router(memories.router, prefix="/memories", tags=["memories"])
-app.include_router(files.router, prefix="/files", tags=["files"])
-app.include_router(tools.router, prefix="/tools", tags=["tools"])
-app.include_router(functions.router, prefix="/functions", tags=["functions"])
-
-app.include_router(utils.router, prefix="/utils", tags=["utils"])
-
-
-@app.get("/")
-async def get_status():
-    return {
-        "status": True,
-        "auth": WEBUI_AUTH,
-        "default_models": app.state.config.DEFAULT_MODELS,
-        "default_prompt_suggestions": app.state.config.DEFAULT_PROMPT_SUGGESTIONS,
-    }
-
-
-async def get_pipe_models():
-    pipes = Functions.get_functions_by_type("pipe", active_only=True)
-    pipe_models = []
-
-    for pipe in pipes:
-        # Check if function is already loaded
-        if pipe.id not in app.state.FUNCTIONS:
-            function_module, function_type, frontmatter = load_function_module_by_id(
-                pipe.id
-            )
-            app.state.FUNCTIONS[pipe.id] = function_module
-        else:
-            function_module = app.state.FUNCTIONS[pipe.id]
-
-        if hasattr(function_module, "valves") and hasattr(function_module, "Valves"):
-            print(f"Getting valves for {pipe.id}")
-            valves = Functions.get_function_valves_by_id(pipe.id)
-            function_module.valves = function_module.Valves(
-                **(valves if valves else {})
-            )
-
-        # Check if function is a manifold
-        if hasattr(function_module, "type"):
-            if function_module.type == "manifold":
-                manifold_pipes = []
-
-                # Check if pipes is a function or a list
-                if callable(function_module.pipes):
-                    manifold_pipes = function_module.pipes()
-                else:
-                    manifold_pipes = function_module.pipes
-
-                for p in manifold_pipes:
-                    manifold_pipe_id = f'{pipe.id}.{p["id"]}'
-                    manifold_pipe_name = p["name"]
-
-                    if hasattr(function_module, "name"):
-                        manifold_pipe_name = (
-                            f"{function_module.name}{manifold_pipe_name}"
-                        )
-
-                    pipe_models.append(
-                        {
-                            "id": manifold_pipe_id,
-                            "name": manifold_pipe_name,
-                            "object": "model",
-                            "created": pipe.created_at,
-                            "owned_by": "openai",
-                            "pipe": {"type": pipe.type},
-                        }
-                    )
-        else:
-            pipe_models.append(
-                {
-                    "id": pipe.id,
-                    "name": pipe.name,
-                    "object": "model",
-                    "created": pipe.created_at,
-                    "owned_by": "openai",
-                    "pipe": {"type": "pipe"},
-                }
-            )
-
-    return pipe_models
-
-
-async def generate_function_chat_completion(form_data, user):
-    async def job():
-        pipe_id = form_data["model"]
-        if "." in pipe_id:
-            pipe_id, sub_pipe_id = pipe_id.split(".", 1)
-        print(pipe_id)
-
-        # Check if function is already loaded
-        if pipe_id not in app.state.FUNCTIONS:
-            function_module, function_type, frontmatter = load_function_module_by_id(
-                pipe_id
-            )
-            app.state.FUNCTIONS[pipe_id] = function_module
-        else:
-            function_module = app.state.FUNCTIONS[pipe_id]
-
-        if hasattr(function_module, "valves") and hasattr(function_module, "Valves"):
-
-            valves = Functions.get_function_valves_by_id(pipe_id)
-            function_module.valves = function_module.Valves(
-                **(valves if valves else {})
-            )
-
-        pipe = function_module.pipe
-
-        # Get the signature of the function
-        sig = inspect.signature(pipe)
-        params = {"body": form_data}
-
-        if "__user__" in sig.parameters:
-            __user__ = {
-                "id": user.id,
-                "email": user.email,
-                "name": user.name,
-                "role": user.role,
-            }
-
-            try:
-                if hasattr(function_module, "UserValves"):
-                    __user__["valves"] = function_module.UserValves(
-                        **Functions.get_user_valves_by_id_and_user_id(pipe_id, user.id)
-                    )
-            except Exception as e:
-                print(e)
-
-            params = {**params, "__user__": __user__}
-
-        if form_data["stream"]:
-
-            async def stream_content():
-                try:
-                    if inspect.iscoroutinefunction(pipe):
-                        res = await pipe(**params)
-                    else:
-                        res = pipe(**params)
-
-                    # Directly return if the response is a StreamingResponse
-                    if isinstance(res, StreamingResponse):
-                        async for data in res.body_iterator:
-                            yield data
-                        return
-                    if isinstance(res, dict):
-                        yield f"data: {json.dumps(res)}\n\n"
-                        return
-
-                except Exception as e:
-                    print(f"Error: {e}")
-                    yield f"data: {json.dumps({'error': {'detail':str(e)}})}\n\n"
-                    return
-
-                if isinstance(res, str):
-                    message = stream_message_template(form_data["model"], res)
-                    yield f"data: {json.dumps(message)}\n\n"
-
-                if isinstance(res, Iterator):
-                    for line in res:
-                        if isinstance(line, BaseModel):
-                            line = line.model_dump_json()
-                            line = f"data: {line}"
-                        try:
-                            line = line.decode("utf-8")
-                        except:
-                            pass
-
-                        if line.startswith("data:"):
-                            yield f"{line}\n\n"
-                        else:
-                            line = stream_message_template(form_data["model"], line)
-                            yield f"data: {json.dumps(line)}\n\n"
-
-                if isinstance(res, str) or isinstance(res, Generator):
-                    finish_message = {
-                        "id": f"{form_data['model']}-{str(uuid.uuid4())}",
-                        "object": "chat.completion.chunk",
-                        "created": int(time.time()),
-                        "model": form_data["model"],
-                        "choices": [
-                            {
-                                "index": 0,
-                                "delta": {},
-                                "logprobs": None,
-                                "finish_reason": "stop",
-                            }
-                        ],
-                    }
-
-                    yield f"data: {json.dumps(finish_message)}\n\n"
-                    yield f"data: [DONE]"
-
-            return StreamingResponse(stream_content(), media_type="text/event-stream")
-        else:
-
-            try:
-                if inspect.iscoroutinefunction(pipe):
-                    res = await pipe(**params)
-                else:
-                    res = pipe(**params)
-
-                if isinstance(res, StreamingResponse):
-                    return res
-            except Exception as e:
-                print(f"Error: {e}")
-                return {"error": {"detail": str(e)}}
-
-            if isinstance(res, dict):
-                return res
-            elif isinstance(res, BaseModel):
-                return res.model_dump()
-            else:
-                message = ""
-                if isinstance(res, str):
-                    message = res
-                if isinstance(res, Generator):
-                    for stream in res:
-                        message = f"{message}{stream}"
-
-                return {
-                    "id": f"{form_data['model']}-{str(uuid.uuid4())}",
-                    "object": "chat.completion",
-                    "created": int(time.time()),
-                    "model": form_data["model"],
-                    "choices": [
-                        {
-                            "index": 0,
-                            "message": {
-                                "role": "assistant",
-                                "content": message,
-                            },
-                            "logprobs": None,
-                            "finish_reason": "stop",
-                        }
-                    ],
-                }
-
-    return await job()

backend/apps/webui/models/chats.py

@@ -1,355 +0,0 @@
-from pydantic import BaseModel
-from typing import List, Union, Optional
-from peewee import *
-from playhouse.shortcuts import model_to_dict
-
-import json
-import uuid
-import time
-
-from apps.webui.internal.db import DB
-
-####################
-# Chat DB Schema
-####################
-
-
-class Chat(Model):
-    id = CharField(unique=True)
-    user_id = CharField()
-    title = TextField()
-    chat = TextField()  # Save Chat JSON as Text
-
-    created_at = BigIntegerField()
-    updated_at = BigIntegerField()
-
-    share_id = CharField(null=True, unique=True)
-    archived = BooleanField(default=False)
-
-    class Meta:
-        database = DB
-
-
-class ChatModel(BaseModel):
-    id: str
-    user_id: str
-    title: str
-    chat: str
-
-    created_at: int  # timestamp in epoch
-    updated_at: int  # timestamp in epoch
-
-    share_id: Optional[str] = None
-    archived: bool = False
-
-
-####################
-# Forms
-####################
-
-
-class ChatForm(BaseModel):
-    chat: dict
-
-
-class ChatTitleForm(BaseModel):
-    title: str
-
-
-class ChatResponse(BaseModel):
-    id: str
-    user_id: str
-    title: str
-    chat: dict
-    updated_at: int  # timestamp in epoch
-    created_at: int  # timestamp in epoch
-    share_id: Optional[str] = None  # id of the chat to be shared
-    archived: bool
-
-
-class ChatTitleIdResponse(BaseModel):
-    id: str
-    title: str
-    updated_at: int
-    created_at: int
-
-
-class ChatTable:
-    def __init__(self, db):
-        self.db = db
-        db.create_tables([Chat])
-
-    def insert_new_chat(self, user_id: str, form_data: ChatForm) -> Optional[ChatModel]:
-        id = str(uuid.uuid4())
-        chat = ChatModel(
-            **{
-                "id": id,
-                "user_id": user_id,
-                "title": (
-                    form_data.chat["title"] if "title" in form_data.chat else "New Chat"
-                ),
-                "chat": json.dumps(form_data.chat),
-                "created_at": int(time.time()),
-                "updated_at": int(time.time()),
-            }
-        )
-
-        result = Chat.create(**chat.model_dump())
-        return chat if result else None
-
-    def update_chat_by_id(self, id: str, chat: dict) -> Optional[ChatModel]:
-        try:
-            query = Chat.update(
-                chat=json.dumps(chat),
-                title=chat["title"] if "title" in chat else "New Chat",
-                updated_at=int(time.time()),
-            ).where(Chat.id == id)
-            query.execute()
-
-            chat = Chat.get(Chat.id == id)
-            return ChatModel(**model_to_dict(chat))
-        except:
-            return None
-
-    def insert_shared_chat_by_chat_id(self, chat_id: str) -> Optional[ChatModel]:
-        # Get the existing chat to share
-        chat = Chat.get(Chat.id == chat_id)
-        # Check if the chat is already shared
-        if chat.share_id:
-            return self.get_chat_by_id_and_user_id(chat.share_id, "shared")
-        # Create a new chat with the same data, but with a new ID
-        shared_chat = ChatModel(
-            **{
-                "id": str(uuid.uuid4()),
-                "user_id": f"shared-{chat_id}",
-                "title": chat.title,
-                "chat": chat.chat,
-                "created_at": chat.created_at,
-                "updated_at": int(time.time()),
-            }
-        )
-        shared_result = Chat.create(**shared_chat.model_dump())
-        # Update the original chat with the share_id
-        result = (
-            Chat.update(share_id=shared_chat.id).where(Chat.id == chat_id).execute()
-        )
-
-        return shared_chat if (shared_result and result) else None
-
-    def update_shared_chat_by_chat_id(self, chat_id: str) -> Optional[ChatModel]:
-        try:
-            print("update_shared_chat_by_id")
-            chat = Chat.get(Chat.id == chat_id)
-            print(chat)
-
-            query = Chat.update(
-                title=chat.title,
-                chat=chat.chat,
-            ).where(Chat.id == chat.share_id)
-
-            query.execute()
-
-            chat = Chat.get(Chat.id == chat.share_id)
-            return ChatModel(**model_to_dict(chat))
-        except:
-            return None
-
-    def delete_shared_chat_by_chat_id(self, chat_id: str) -> bool:
-        try:
-            query = Chat.delete().where(Chat.user_id == f"shared-{chat_id}")
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except:
-            return False
-
-    def update_chat_share_id_by_id(
-        self, id: str, share_id: Optional[str]
-    ) -> Optional[ChatModel]:
-        try:
-            query = Chat.update(
-                share_id=share_id,
-            ).where(Chat.id == id)
-            query.execute()
-
-            chat = Chat.get(Chat.id == id)
-            return ChatModel(**model_to_dict(chat))
-        except:
-            return None
-
-    def toggle_chat_archive_by_id(self, id: str) -> Optional[ChatModel]:
-        try:
-            chat = self.get_chat_by_id(id)
-            query = Chat.update(
-                archived=(not chat.archived),
-            ).where(Chat.id == id)
-
-            query.execute()
-
-            chat = Chat.get(Chat.id == id)
-            return ChatModel(**model_to_dict(chat))
-        except:
-            return None
-
-    def archive_all_chats_by_user_id(self, user_id: str) -> bool:
-        try:
-            chats = self.get_chats_by_user_id(user_id)
-            for chat in chats:
-                query = Chat.update(
-                    archived=True,
-                ).where(Chat.id == chat.id)
-
-                query.execute()
-
-            return True
-        except:
-            return False
-
-    def get_archived_chat_list_by_user_id(
-        self, user_id: str, skip: int = 0, limit: int = 50
-    ) -> List[ChatModel]:
-        return [
-            ChatModel(**model_to_dict(chat))
-            for chat in Chat.select()
-            .where(Chat.archived == True)
-            .where(Chat.user_id == user_id)
-            .order_by(Chat.updated_at.desc())
-            # .limit(limit)
-            # .offset(skip)
-        ]
-
-    def get_chat_list_by_user_id(
-        self,
-        user_id: str,
-        include_archived: bool = False,
-        skip: int = 0,
-        limit: int = 50,
-    ) -> List[ChatModel]:
-        if include_archived:
-            return [
-                ChatModel(**model_to_dict(chat))
-                for chat in Chat.select()
-                .where(Chat.user_id == user_id)
-                .order_by(Chat.updated_at.desc())
-                # .limit(limit)
-                # .offset(skip)
-            ]
-        else:
-            return [
-                ChatModel(**model_to_dict(chat))
-                for chat in Chat.select()
-                .where(Chat.archived == False)
-                .where(Chat.user_id == user_id)
-                .order_by(Chat.updated_at.desc())
-                # .limit(limit)
-                # .offset(skip)
-            ]
-
-    def get_chat_list_by_chat_ids(
-        self, chat_ids: List[str], skip: int = 0, limit: int = 50
-    ) -> List[ChatModel]:
-        return [
-            ChatModel(**model_to_dict(chat))
-            for chat in Chat.select()
-            .where(Chat.archived == False)
-            .where(Chat.id.in_(chat_ids))
-            .order_by(Chat.updated_at.desc())
-        ]
-
-    def get_chat_by_id(self, id: str) -> Optional[ChatModel]:
-        try:
-            chat = Chat.get(Chat.id == id)
-            return ChatModel(**model_to_dict(chat))
-        except:
-            return None
-
-    def get_chat_by_share_id(self, id: str) -> Optional[ChatModel]:
-        try:
-            chat = Chat.get(Chat.share_id == id)
-
-            if chat:
-                chat = Chat.get(Chat.id == id)
-                return ChatModel(**model_to_dict(chat))
-            else:
-                return None
-        except:
-            return None
-
-    def get_chat_by_id_and_user_id(self, id: str, user_id: str) -> Optional[ChatModel]:
-        try:
-            chat = Chat.get(Chat.id == id, Chat.user_id == user_id)
-            return ChatModel(**model_to_dict(chat))
-        except:
-            return None
-
-    def get_chats(self, skip: int = 0, limit: int = 50) -> List[ChatModel]:
-        return [
-            ChatModel(**model_to_dict(chat))
-            for chat in Chat.select().order_by(Chat.updated_at.desc())
-            # .limit(limit).offset(skip)
-        ]
-
-    def get_chats_by_user_id(self, user_id: str) -> List[ChatModel]:
-        return [
-            ChatModel(**model_to_dict(chat))
-            for chat in Chat.select()
-            .where(Chat.user_id == user_id)
-            .order_by(Chat.updated_at.desc())
-            # .limit(limit).offset(skip)
-        ]
-
-    def get_archived_chats_by_user_id(self, user_id: str) -> List[ChatModel]:
-        return [
-            ChatModel(**model_to_dict(chat))
-            for chat in Chat.select()
-            .where(Chat.archived == True)
-            .where(Chat.user_id == user_id)
-            .order_by(Chat.updated_at.desc())
-        ]
-
-    def delete_chat_by_id(self, id: str) -> bool:
-        try:
-            query = Chat.delete().where((Chat.id == id))
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True and self.delete_shared_chat_by_chat_id(id)
-        except:
-            return False
-
-    def delete_chat_by_id_and_user_id(self, id: str, user_id: str) -> bool:
-        try:
-            query = Chat.delete().where((Chat.id == id) & (Chat.user_id == user_id))
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True and self.delete_shared_chat_by_chat_id(id)
-        except:
-            return False
-
-    def delete_chats_by_user_id(self, user_id: str) -> bool:
-        try:
-
-            self.delete_shared_chats_by_user_id(user_id)
-
-            query = Chat.delete().where(Chat.user_id == user_id)
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except:
-            return False
-
-    def delete_shared_chats_by_user_id(self, user_id: str) -> bool:
-        try:
-            shared_chat_ids = [
-                f"shared-{chat.id}"
-                for chat in Chat.select().where(Chat.user_id == user_id)
-            ]
-
-            query = Chat.delete().where(Chat.user_id << shared_chat_ids)
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except:
-            return False
-
-
-Chats = ChatTable(DB)

backend/apps/webui/models/documents.py

@@ -1,160 +0,0 @@
-from pydantic import BaseModel
-from peewee import *
-from playhouse.shortcuts import model_to_dict
-from typing import List, Union, Optional
-import time
-import logging
-
-from utils.utils import decode_token
-from utils.misc import get_gravatar_url
-
-from apps.webui.internal.db import DB
-
-import json
-
-from config import SRC_LOG_LEVELS
-
-log = logging.getLogger(__name__)
-log.setLevel(SRC_LOG_LEVELS["MODELS"])
-
-####################
-# Documents DB Schema
-####################
-
-
-class Document(Model):
-    collection_name = CharField(unique=True)
-    name = CharField(unique=True)
-    title = TextField()
-    filename = TextField()
-    content = TextField(null=True)
-    user_id = CharField()
-    timestamp = BigIntegerField()
-
-    class Meta:
-        database = DB
-
-
-class DocumentModel(BaseModel):
-    collection_name: str
-    name: str
-    title: str
-    filename: str
-    content: Optional[str] = None
-    user_id: str
-    timestamp: int  # timestamp in epoch
-
-
-####################
-# Forms
-####################
-
-
-class DocumentResponse(BaseModel):
-    collection_name: str
-    name: str
-    title: str
-    filename: str
-    content: Optional[dict] = None
-    user_id: str
-    timestamp: int  # timestamp in epoch
-
-
-class DocumentUpdateForm(BaseModel):
-    name: str
-    title: str
-
-
-class DocumentForm(DocumentUpdateForm):
-    collection_name: str
-    filename: str
-    content: Optional[str] = None
-
-
-class DocumentsTable:
-    def __init__(self, db):
-        self.db = db
-        self.db.create_tables([Document])
-
-    def insert_new_doc(
-        self, user_id: str, form_data: DocumentForm
-    ) -> Optional[DocumentModel]:
-        document = DocumentModel(
-            **{
-                **form_data.model_dump(),
-                "user_id": user_id,
-                "timestamp": int(time.time()),
-            }
-        )
-
-        try:
-            result = Document.create(**document.model_dump())
-            if result:
-                return document
-            else:
-                return None
-        except:
-            return None
-
-    def get_doc_by_name(self, name: str) -> Optional[DocumentModel]:
-        try:
-            document = Document.get(Document.name == name)
-            return DocumentModel(**model_to_dict(document))
-        except:
-            return None
-
-    def get_docs(self) -> List[DocumentModel]:
-        return [
-            DocumentModel(**model_to_dict(doc))
-            for doc in Document.select()
-            # .limit(limit).offset(skip)
-        ]
-
-    def update_doc_by_name(
-        self, name: str, form_data: DocumentUpdateForm
-    ) -> Optional[DocumentModel]:
-        try:
-            query = Document.update(
-                title=form_data.title,
-                name=form_data.name,
-                timestamp=int(time.time()),
-            ).where(Document.name == name)
-            query.execute()
-
-            doc = Document.get(Document.name == form_data.name)
-            return DocumentModel(**model_to_dict(doc))
-        except Exception as e:
-            log.exception(e)
-            return None
-
-    def update_doc_content_by_name(
-        self, name: str, updated: dict
-    ) -> Optional[DocumentModel]:
-        try:
-            doc = self.get_doc_by_name(name)
-            doc_content = json.loads(doc.content if doc.content else "{}")
-            doc_content = {**doc_content, **updated}
-
-            query = Document.update(
-                content=json.dumps(doc_content),
-                timestamp=int(time.time()),
-            ).where(Document.name == name)
-            query.execute()
-
-            doc = Document.get(Document.name == name)
-            return DocumentModel(**model_to_dict(doc))
-        except Exception as e:
-            log.exception(e)
-            return None
-
-    def delete_doc_by_name(self, name: str) -> bool:
-        try:
-            query = Document.delete().where((Document.name == name))
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except:
-            return False
-
-
-Documents = DocumentsTable(DB)

backend/apps/webui/models/files.py

@@ -1,112 +0,0 @@
-from pydantic import BaseModel
-from peewee import *
-from playhouse.shortcuts import model_to_dict
-from typing import List, Union, Optional
-import time
-import logging
-from apps.webui.internal.db import DB, JSONField
-
-import json
-
-from config import SRC_LOG_LEVELS
-
-log = logging.getLogger(__name__)
-log.setLevel(SRC_LOG_LEVELS["MODELS"])
-
-####################
-# Files DB Schema
-####################
-
-
-class File(Model):
-    id = CharField(unique=True)
-    user_id = CharField()
-    filename = TextField()
-    meta = JSONField()
-    created_at = BigIntegerField()
-
-    class Meta:
-        database = DB
-
-
-class FileModel(BaseModel):
-    id: str
-    user_id: str
-    filename: str
-    meta: dict
-    created_at: int  # timestamp in epoch
-
-
-####################
-# Forms
-####################
-
-
-class FileModelResponse(BaseModel):
-    id: str
-    user_id: str
-    filename: str
-    meta: dict
-    created_at: int  # timestamp in epoch
-
-
-class FileForm(BaseModel):
-    id: str
-    filename: str
-    meta: dict = {}
-
-
-class FilesTable:
-    def __init__(self, db):
-        self.db = db
-        self.db.create_tables([File])
-
-    def insert_new_file(self, user_id: str, form_data: FileForm) -> Optional[FileModel]:
-        file = FileModel(
-            **{
-                **form_data.model_dump(),
-                "user_id": user_id,
-                "created_at": int(time.time()),
-            }
-        )
-
-        try:
-            result = File.create(**file.model_dump())
-            if result:
-                return file
-            else:
-                return None
-        except Exception as e:
-            print(f"Error creating tool: {e}")
-            return None
-
-    def get_file_by_id(self, id: str) -> Optional[FileModel]:
-        try:
-            file = File.get(File.id == id)
-            return FileModel(**model_to_dict(file))
-        except:
-            return None
-
-    def get_files(self) -> List[FileModel]:
-        return [FileModel(**model_to_dict(file)) for file in File.select()]
-
-    def delete_file_by_id(self, id: str) -> bool:
-        try:
-            query = File.delete().where((File.id == id))
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except:
-            return False
-
-    def delete_all_files(self) -> bool:
-        try:
-            query = File.delete()
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except:
-            return False
-
-
-Files = FilesTable(DB)

backend/apps/webui/models/functions.py

@@ -1,261 +0,0 @@
-from pydantic import BaseModel
-from peewee import *
-from playhouse.shortcuts import model_to_dict
-from typing import List, Union, Optional
-import time
-import logging
-from apps.webui.internal.db import DB, JSONField
-from apps.webui.models.users import Users
-
-import json
-import copy
-
-
-from config import SRC_LOG_LEVELS
-
-log = logging.getLogger(__name__)
-log.setLevel(SRC_LOG_LEVELS["MODELS"])
-
-####################
-# Functions DB Schema
-####################
-
-
-class Function(Model):
-    id = CharField(unique=True)
-    user_id = CharField()
-    name = TextField()
-    type = TextField()
-    content = TextField()
-    meta = JSONField()
-    valves = JSONField()
-    is_active = BooleanField(default=False)
-    is_global = BooleanField(default=False)
-    updated_at = BigIntegerField()
-    created_at = BigIntegerField()
-
-    class Meta:
-        database = DB
-
-
-class FunctionMeta(BaseModel):
-    description: Optional[str] = None
-    manifest: Optional[dict] = {}
-
-
-class FunctionModel(BaseModel):
-    id: str
-    user_id: str
-    name: str
-    type: str
-    content: str
-    meta: FunctionMeta
-    is_active: bool = False
-    is_global: bool = False
-    updated_at: int  # timestamp in epoch
-    created_at: int  # timestamp in epoch
-
-
-####################
-# Forms
-####################
-
-
-class FunctionResponse(BaseModel):
-    id: str
-    user_id: str
-    type: str
-    name: str
-    meta: FunctionMeta
-    is_active: bool
-    is_global: bool
-    updated_at: int  # timestamp in epoch
-    created_at: int  # timestamp in epoch
-
-
-class FunctionForm(BaseModel):
-    id: str
-    name: str
-    content: str
-    meta: FunctionMeta
-
-
-class FunctionValves(BaseModel):
-    valves: Optional[dict] = None
-
-
-class FunctionsTable:
-    def __init__(self, db):
-        self.db = db
-        self.db.create_tables([Function])
-
-    def insert_new_function(
-        self, user_id: str, type: str, form_data: FunctionForm
-    ) -> Optional[FunctionModel]:
-        function = FunctionModel(
-            **{
-                **form_data.model_dump(),
-                "user_id": user_id,
-                "type": type,
-                "updated_at": int(time.time()),
-                "created_at": int(time.time()),
-            }
-        )
-
-        try:
-            result = Function.create(**function.model_dump())
-            if result:
-                return function
-            else:
-                return None
-        except Exception as e:
-            print(f"Error creating tool: {e}")
-            return None
-
-    def get_function_by_id(self, id: str) -> Optional[FunctionModel]:
-        try:
-            function = Function.get(Function.id == id)
-            return FunctionModel(**model_to_dict(function))
-        except:
-            return None
-
-    def get_functions(self, active_only=False) -> List[FunctionModel]:
-        if active_only:
-            return [
-                FunctionModel(**model_to_dict(function))
-                for function in Function.select().where(Function.is_active == True)
-            ]
-        else:
-            return [
-                FunctionModel(**model_to_dict(function))
-                for function in Function.select()
-            ]
-
-    def get_functions_by_type(
-        self, type: str, active_only=False
-    ) -> List[FunctionModel]:
-        if active_only:
-            return [
-                FunctionModel(**model_to_dict(function))
-                for function in Function.select().where(
-                    Function.type == type, Function.is_active == True
-                )
-            ]
-        else:
-            return [
-                FunctionModel(**model_to_dict(function))
-                for function in Function.select().where(Function.type == type)
-            ]
-
-    def get_global_filter_functions(self) -> List[FunctionModel]:
-        return [
-            FunctionModel(**model_to_dict(function))
-            for function in Function.select().where(
-                Function.type == "filter",
-                Function.is_active == True,
-                Function.is_global == True,
-            )
-        ]
-
-    def get_function_valves_by_id(self, id: str) -> Optional[dict]:
-        try:
-            function = Function.get(Function.id == id)
-            return function.valves if function.valves else {}
-        except Exception as e:
-            print(f"An error occurred: {e}")
-            return None
-
-    def update_function_valves_by_id(
-        self, id: str, valves: dict
-    ) -> Optional[FunctionValves]:
-        try:
-            query = Function.update(
-                **{"valves": valves},
-                updated_at=int(time.time()),
-            ).where(Function.id == id)
-            query.execute()
-
-            function = Function.get(Function.id == id)
-            return FunctionValves(**model_to_dict(function))
-        except:
-            return None
-
-    def get_user_valves_by_id_and_user_id(
-        self, id: str, user_id: str
-    ) -> Optional[dict]:
-        try:
-            user = Users.get_user_by_id(user_id)
-            user_settings = user.settings.model_dump()
-
-            # Check if user has "functions" and "valves" settings
-            if "functions" not in user_settings:
-                user_settings["functions"] = {}
-            if "valves" not in user_settings["functions"]:
-                user_settings["functions"]["valves"] = {}
-
-            return user_settings["functions"]["valves"].get(id, {})
-        except Exception as e:
-            print(f"An error occurred: {e}")
-            return None
-
-    def update_user_valves_by_id_and_user_id(
-        self, id: str, user_id: str, valves: dict
-    ) -> Optional[dict]:
-        try:
-            user = Users.get_user_by_id(user_id)
-            user_settings = user.settings.model_dump()
-
-            # Check if user has "functions" and "valves" settings
-            if "functions" not in user_settings:
-                user_settings["functions"] = {}
-            if "valves" not in user_settings["functions"]:
-                user_settings["functions"]["valves"] = {}
-
-            user_settings["functions"]["valves"][id] = valves
-
-            # Update the user settings in the database
-            query = Users.update_user_by_id(user_id, {"settings": user_settings})
-            query.execute()
-
-            return user_settings["functions"]["valves"][id]
-        except Exception as e:
-            print(f"An error occurred: {e}")
-            return None
-
-    def update_function_by_id(self, id: str, updated: dict) -> Optional[FunctionModel]:
-        try:
-            query = Function.update(
-                **updated,
-                updated_at=int(time.time()),
-            ).where(Function.id == id)
-            query.execute()
-
-            function = Function.get(Function.id == id)
-            return FunctionModel(**model_to_dict(function))
-        except:
-            return None
-
-    def deactivate_all_functions(self) -> Optional[bool]:
-        try:
-            query = Function.update(
-                **{"is_active": False},
-                updated_at=int(time.time()),
-            )
-
-            query.execute()
-
-            return True
-        except:
-            return None
-
-    def delete_function_by_id(self, id: str) -> bool:
-        try:
-            query = Function.delete().where((Function.id == id))
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except:
-            return False
-
-
-Functions = FunctionsTable(DB)

backend/apps/webui/models/memories.py

@@ -1,132 +0,0 @@
-from pydantic import BaseModel
-from peewee import *
-from playhouse.shortcuts import model_to_dict
-from typing import List, Union, Optional
-
-from apps.webui.internal.db import DB
-from apps.webui.models.chats import Chats
-
-import time
-import uuid
-
-####################
-# Memory DB Schema
-####################
-
-
-class Memory(Model):
-    id = CharField(unique=True)
-    user_id = CharField()
-    content = TextField()
-    updated_at = BigIntegerField()
-    created_at = BigIntegerField()
-
-    class Meta:
-        database = DB
-
-
-class MemoryModel(BaseModel):
-    id: str
-    user_id: str
-    content: str
-    updated_at: int  # timestamp in epoch
-    created_at: int  # timestamp in epoch
-
-
-####################
-# Forms
-####################
-
-
-class MemoriesTable:
-    def __init__(self, db):
-        self.db = db
-        self.db.create_tables([Memory])
-
-    def insert_new_memory(
-        self,
-        user_id: str,
-        content: str,
-    ) -> Optional[MemoryModel]:
-        id = str(uuid.uuid4())
-
-        memory = MemoryModel(
-            **{
-                "id": id,
-                "user_id": user_id,
-                "content": content,
-                "created_at": int(time.time()),
-                "updated_at": int(time.time()),
-            }
-        )
-        result = Memory.create(**memory.model_dump())
-        if result:
-            return memory
-        else:
-            return None
-
-    def update_memory_by_id(
-        self,
-        id: str,
-        content: str,
-    ) -> Optional[MemoryModel]:
-        try:
-            memory = Memory.get(Memory.id == id)
-            memory.content = content
-            memory.updated_at = int(time.time())
-            memory.save()
-            return MemoryModel(**model_to_dict(memory))
-        except:
-            return None
-
-    def get_memories(self) -> List[MemoryModel]:
-        try:
-            memories = Memory.select()
-            return [MemoryModel(**model_to_dict(memory)) for memory in memories]
-        except:
-            return None
-
-    def get_memories_by_user_id(self, user_id: str) -> List[MemoryModel]:
-        try:
-            memories = Memory.select().where(Memory.user_id == user_id)
-            return [MemoryModel(**model_to_dict(memory)) for memory in memories]
-        except:
-            return None
-
-    def get_memory_by_id(self, id) -> Optional[MemoryModel]:
-        try:
-            memory = Memory.get(Memory.id == id)
-            return MemoryModel(**model_to_dict(memory))
-        except:
-            return None
-
-    def delete_memory_by_id(self, id: str) -> bool:
-        try:
-            query = Memory.delete().where(Memory.id == id)
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-
-        except:
-            return False
-
-    def delete_memories_by_user_id(self, user_id: str) -> bool:
-        try:
-            query = Memory.delete().where(Memory.user_id == user_id)
-            query.execute()
-
-            return True
-        except:
-            return False
-
-    def delete_memory_by_id_and_user_id(self, id: str, user_id: str) -> bool:
-        try:
-            query = Memory.delete().where(Memory.id == id, Memory.user_id == user_id)
-            query.execute()
-
-            return True
-        except:
-            return False
-
-
-Memories = MemoriesTable(DB)

backend/apps/webui/models/prompts.py

@@ -1,118 +0,0 @@
-from pydantic import BaseModel
-from peewee import *
-from playhouse.shortcuts import model_to_dict
-from typing import List, Union, Optional
-import time
-
-from utils.utils import decode_token
-from utils.misc import get_gravatar_url
-
-from apps.webui.internal.db import DB
-
-import json
-
-####################
-# Prompts DB Schema
-####################
-
-
-class Prompt(Model):
-    command = CharField(unique=True)
-    user_id = CharField()
-    title = TextField()
-    content = TextField()
-    timestamp = BigIntegerField()
-
-    class Meta:
-        database = DB
-
-
-class PromptModel(BaseModel):
-    command: str
-    user_id: str
-    title: str
-    content: str
-    timestamp: int  # timestamp in epoch
-
-
-####################
-# Forms
-####################
-
-
-class PromptForm(BaseModel):
-    command: str
-    title: str
-    content: str
-
-
-class PromptsTable:
-
-    def __init__(self, db):
-        self.db = db
-        self.db.create_tables([Prompt])
-
-    def insert_new_prompt(
-        self, user_id: str, form_data: PromptForm
-    ) -> Optional[PromptModel]:
-        prompt = PromptModel(
-            **{
-                "user_id": user_id,
-                "command": form_data.command,
-                "title": form_data.title,
-                "content": form_data.content,
-                "timestamp": int(time.time()),
-            }
-        )
-
-        try:
-            result = Prompt.create(**prompt.model_dump())
-            if result:
-                return prompt
-            else:
-                return None
-        except:
-            return None
-
-    def get_prompt_by_command(self, command: str) -> Optional[PromptModel]:
-        try:
-            prompt = Prompt.get(Prompt.command == command)
-            return PromptModel(**model_to_dict(prompt))
-        except:
-            return None
-
-    def get_prompts(self) -> List[PromptModel]:
-        return [
-            PromptModel(**model_to_dict(prompt))
-            for prompt in Prompt.select()
-            # .limit(limit).offset(skip)
-        ]
-
-    def update_prompt_by_command(
-        self, command: str, form_data: PromptForm
-    ) -> Optional[PromptModel]:
-        try:
-            query = Prompt.update(
-                title=form_data.title,
-                content=form_data.content,
-                timestamp=int(time.time()),
-            ).where(Prompt.command == command)
-
-            query.execute()
-
-            prompt = Prompt.get(Prompt.command == command)
-            return PromptModel(**model_to_dict(prompt))
-        except:
-            return None
-
-    def delete_prompt_by_command(self, command: str) -> bool:
-        try:
-            query = Prompt.delete().where((Prompt.command == command))
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except:
-            return False
-
-
-Prompts = PromptsTable(DB)

backend/apps/webui/models/tags.py

@@ -1,237 +0,0 @@
-from pydantic import BaseModel
-from typing import List, Union, Optional
-from peewee import *
-from playhouse.shortcuts import model_to_dict
-
-import json
-import uuid
-import time
-import logging
-
-from apps.webui.internal.db import DB
-
-from config import SRC_LOG_LEVELS
-
-log = logging.getLogger(__name__)
-log.setLevel(SRC_LOG_LEVELS["MODELS"])
-
-####################
-# Tag DB Schema
-####################
-
-
-class Tag(Model):
-    id = CharField(unique=True)
-    name = CharField()
-    user_id = CharField()
-    data = TextField(null=True)
-
-    class Meta:
-        database = DB
-
-
-class ChatIdTag(Model):
-    id = CharField(unique=True)
-    tag_name = CharField()
-    chat_id = CharField()
-    user_id = CharField()
-    timestamp = BigIntegerField()
-
-    class Meta:
-        database = DB
-
-
-class TagModel(BaseModel):
-    id: str
-    name: str
-    user_id: str
-    data: Optional[str] = None
-
-
-class ChatIdTagModel(BaseModel):
-    id: str
-    tag_name: str
-    chat_id: str
-    user_id: str
-    timestamp: int
-
-
-####################
-# Forms
-####################
-
-
-class ChatIdTagForm(BaseModel):
-    tag_name: str
-    chat_id: str
-
-
-class TagChatIdsResponse(BaseModel):
-    chat_ids: List[str]
-
-
-class ChatTagsResponse(BaseModel):
-    tags: List[str]
-
-
-class TagTable:
-    def __init__(self, db):
-        self.db = db
-        db.create_tables([Tag, ChatIdTag])
-
-    def insert_new_tag(self, name: str, user_id: str) -> Optional[TagModel]:
-        id = str(uuid.uuid4())
-        tag = TagModel(**{"id": id, "user_id": user_id, "name": name})
-        try:
-            result = Tag.create(**tag.model_dump())
-            if result:
-                return tag
-            else:
-                return None
-        except Exception as e:
-            return None
-
-    def get_tag_by_name_and_user_id(
-        self, name: str, user_id: str
-    ) -> Optional[TagModel]:
-        try:
-            tag = Tag.get(Tag.name == name, Tag.user_id == user_id)
-            return TagModel(**model_to_dict(tag))
-        except Exception as e:
-            return None
-
-    def add_tag_to_chat(
-        self, user_id: str, form_data: ChatIdTagForm
-    ) -> Optional[ChatIdTagModel]:
-        tag = self.get_tag_by_name_and_user_id(form_data.tag_name, user_id)
-        if tag == None:
-            tag = self.insert_new_tag(form_data.tag_name, user_id)
-
-        id = str(uuid.uuid4())
-        chatIdTag = ChatIdTagModel(
-            **{
-                "id": id,
-                "user_id": user_id,
-                "chat_id": form_data.chat_id,
-                "tag_name": tag.name,
-                "timestamp": int(time.time()),
-            }
-        )
-        try:
-            result = ChatIdTag.create(**chatIdTag.model_dump())
-            if result:
-                return chatIdTag
-            else:
-                return None
-        except:
-            return None
-
-    def get_tags_by_user_id(self, user_id: str) -> List[TagModel]:
-        tag_names = [
-            ChatIdTagModel(**model_to_dict(chat_id_tag)).tag_name
-            for chat_id_tag in ChatIdTag.select()
-            .where(ChatIdTag.user_id == user_id)
-            .order_by(ChatIdTag.timestamp.desc())
-        ]
-
-        return [
-            TagModel(**model_to_dict(tag))
-            for tag in Tag.select()
-            .where(Tag.user_id == user_id)
-            .where(Tag.name.in_(tag_names))
-        ]
-
-    def get_tags_by_chat_id_and_user_id(
-        self, chat_id: str, user_id: str
-    ) -> List[TagModel]:
-        tag_names = [
-            ChatIdTagModel(**model_to_dict(chat_id_tag)).tag_name
-            for chat_id_tag in ChatIdTag.select()
-            .where((ChatIdTag.user_id == user_id) & (ChatIdTag.chat_id == chat_id))
-            .order_by(ChatIdTag.timestamp.desc())
-        ]
-
-        return [
-            TagModel(**model_to_dict(tag))
-            for tag in Tag.select()
-            .where(Tag.user_id == user_id)
-            .where(Tag.name.in_(tag_names))
-        ]
-
-    def get_chat_ids_by_tag_name_and_user_id(
-        self, tag_name: str, user_id: str
-    ) -> Optional[ChatIdTagModel]:
-        return [
-            ChatIdTagModel(**model_to_dict(chat_id_tag))
-            for chat_id_tag in ChatIdTag.select()
-            .where((ChatIdTag.user_id == user_id) & (ChatIdTag.tag_name == tag_name))
-            .order_by(ChatIdTag.timestamp.desc())
-        ]
-
-    def count_chat_ids_by_tag_name_and_user_id(
-        self, tag_name: str, user_id: str
-    ) -> int:
-        return (
-            ChatIdTag.select()
-            .where((ChatIdTag.tag_name == tag_name) & (ChatIdTag.user_id == user_id))
-            .count()
-        )
-
-    def delete_tag_by_tag_name_and_user_id(self, tag_name: str, user_id: str) -> bool:
-        try:
-            query = ChatIdTag.delete().where(
-                (ChatIdTag.tag_name == tag_name) & (ChatIdTag.user_id == user_id)
-            )
-            res = query.execute()  # Remove the rows, return number of rows removed.
-            log.debug(f"res: {res}")
-
-            tag_count = self.count_chat_ids_by_tag_name_and_user_id(tag_name, user_id)
-            if tag_count == 0:
-                # Remove tag item from Tag col as well
-                query = Tag.delete().where(
-                    (Tag.name == tag_name) & (Tag.user_id == user_id)
-                )
-                query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except Exception as e:
-            log.error(f"delete_tag: {e}")
-            return False
-
-    def delete_tag_by_tag_name_and_chat_id_and_user_id(
-        self, tag_name: str, chat_id: str, user_id: str
-    ) -> bool:
-        try:
-            query = ChatIdTag.delete().where(
-                (ChatIdTag.tag_name == tag_name)
-                & (ChatIdTag.chat_id == chat_id)
-                & (ChatIdTag.user_id == user_id)
-            )
-            res = query.execute()  # Remove the rows, return number of rows removed.
-            log.debug(f"res: {res}")
-
-            tag_count = self.count_chat_ids_by_tag_name_and_user_id(tag_name, user_id)
-            if tag_count == 0:
-                # Remove tag item from Tag col as well
-                query = Tag.delete().where(
-                    (Tag.name == tag_name) & (Tag.user_id == user_id)
-                )
-                query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except Exception as e:
-            log.error(f"delete_tag: {e}")
-            return False
-
-    def delete_tags_by_chat_id_and_user_id(self, chat_id: str, user_id: str) -> bool:
-        tags = self.get_tags_by_chat_id_and_user_id(chat_id, user_id)
-
-        for tag in tags:
-            self.delete_tag_by_tag_name_and_chat_id_and_user_id(
-                tag.tag_name, chat_id, user_id
-            )
-
-        return True
-
-
-Tags = TagTable(DB)

backend/apps/webui/models/tools.py

@@ -1,204 +0,0 @@
-from pydantic import BaseModel
-from peewee import *
-from playhouse.shortcuts import model_to_dict
-from typing import List, Union, Optional
-import time
-import logging
-from apps.webui.internal.db import DB, JSONField
-from apps.webui.models.users import Users
-
-import json
-import copy
-
-
-from config import SRC_LOG_LEVELS
-
-log = logging.getLogger(__name__)
-log.setLevel(SRC_LOG_LEVELS["MODELS"])
-
-####################
-# Tools DB Schema
-####################
-
-
-class Tool(Model):
-    id = CharField(unique=True)
-    user_id = CharField()
-    name = TextField()
-    content = TextField()
-    specs = JSONField()
-    meta = JSONField()
-    valves = JSONField()
-    updated_at = BigIntegerField()
-    created_at = BigIntegerField()
-
-    class Meta:
-        database = DB
-
-
-class ToolMeta(BaseModel):
-    description: Optional[str] = None
-    manifest: Optional[dict] = {}
-
-
-class ToolModel(BaseModel):
-    id: str
-    user_id: str
-    name: str
-    content: str
-    specs: List[dict]
-    meta: ToolMeta
-    updated_at: int  # timestamp in epoch
-    created_at: int  # timestamp in epoch
-
-
-####################
-# Forms
-####################
-
-
-class ToolResponse(BaseModel):
-    id: str
-    user_id: str
-    name: str
-    meta: ToolMeta
-    updated_at: int  # timestamp in epoch
-    created_at: int  # timestamp in epoch
-
-
-class ToolForm(BaseModel):
-    id: str
-    name: str
-    content: str
-    meta: ToolMeta
-
-
-class ToolValves(BaseModel):
-    valves: Optional[dict] = None
-
-
-class ToolsTable:
-    def __init__(self, db):
-        self.db = db
-        self.db.create_tables([Tool])
-
-    def insert_new_tool(
-        self, user_id: str, form_data: ToolForm, specs: List[dict]
-    ) -> Optional[ToolModel]:
-        tool = ToolModel(
-            **{
-                **form_data.model_dump(),
-                "specs": specs,
-                "user_id": user_id,
-                "updated_at": int(time.time()),
-                "created_at": int(time.time()),
-            }
-        )
-
-        try:
-            result = Tool.create(**tool.model_dump())
-            if result:
-                return tool
-            else:
-                return None
-        except Exception as e:
-            print(f"Error creating tool: {e}")
-            return None
-
-    def get_tool_by_id(self, id: str) -> Optional[ToolModel]:
-        try:
-            tool = Tool.get(Tool.id == id)
-            return ToolModel(**model_to_dict(tool))
-        except:
-            return None
-
-    def get_tools(self) -> List[ToolModel]:
-        return [ToolModel(**model_to_dict(tool)) for tool in Tool.select()]
-
-    def get_tool_valves_by_id(self, id: str) -> Optional[dict]:
-        try:
-            tool = Tool.get(Tool.id == id)
-            return tool.valves if tool.valves else {}
-        except Exception as e:
-            print(f"An error occurred: {e}")
-            return None
-
-    def update_tool_valves_by_id(self, id: str, valves: dict) -> Optional[ToolValves]:
-        try:
-            query = Tool.update(
-                **{"valves": valves},
-                updated_at=int(time.time()),
-            ).where(Tool.id == id)
-            query.execute()
-
-            tool = Tool.get(Tool.id == id)
-            return ToolValves(**model_to_dict(tool))
-        except:
-            return None
-
-    def get_user_valves_by_id_and_user_id(
-        self, id: str, user_id: str
-    ) -> Optional[dict]:
-        try:
-            user = Users.get_user_by_id(user_id)
-            user_settings = user.settings.model_dump()
-
-            # Check if user has "tools" and "valves" settings
-            if "tools" not in user_settings:
-                user_settings["tools"] = {}
-            if "valves" not in user_settings["tools"]:
-                user_settings["tools"]["valves"] = {}
-
-            return user_settings["tools"]["valves"].get(id, {})
-        except Exception as e:
-            print(f"An error occurred: {e}")
-            return None
-
-    def update_user_valves_by_id_and_user_id(
-        self, id: str, user_id: str, valves: dict
-    ) -> Optional[dict]:
-        try:
-            user = Users.get_user_by_id(user_id)
-            user_settings = user.settings.model_dump()
-
-            # Check if user has "tools" and "valves" settings
-            if "tools" not in user_settings:
-                user_settings["tools"] = {}
-            if "valves" not in user_settings["tools"]:
-                user_settings["tools"]["valves"] = {}
-
-            user_settings["tools"]["valves"][id] = valves
-
-            # Update the user settings in the database
-            query = Users.update_user_by_id(user_id, {"settings": user_settings})
-            query.execute()
-
-            return user_settings["tools"]["valves"][id]
-        except Exception as e:
-            print(f"An error occurred: {e}")
-            return None
-
-    def update_tool_by_id(self, id: str, updated: dict) -> Optional[ToolModel]:
-        try:
-            query = Tool.update(
-                **updated,
-                updated_at=int(time.time()),
-            ).where(Tool.id == id)
-            query.execute()
-
-            tool = Tool.get(Tool.id == id)
-            return ToolModel(**model_to_dict(tool))
-        except:
-            return None
-
-    def delete_tool_by_id(self, id: str) -> bool:
-        try:
-            query = Tool.delete().where((Tool.id == id))
-            query.execute()  # Remove the rows, return number of rows removed.
-
-            return True
-        except:
-            return False
-
-
-Tools = ToolsTable(DB)

backend/apps/webui/models/users.py

@@ -1,245 +0,0 @@
-from pydantic import BaseModel, ConfigDict
-from peewee import *
-from playhouse.shortcuts import model_to_dict
-from typing import List, Union, Optional
-import time
-from utils.misc import get_gravatar_url
-
-from apps.webui.internal.db import DB, JSONField
-from apps.webui.models.chats import Chats
-
-####################
-# User DB Schema
-####################
-
-
-class User(Model):
-    id = CharField(unique=True)
-    name = CharField()
-    email = CharField()
-    role = CharField()
-    profile_image_url = TextField()
-
-    last_active_at = BigIntegerField()
-    updated_at = BigIntegerField()
-    created_at = BigIntegerField()
-
-    api_key = CharField(null=True, unique=True)
-    settings = JSONField(null=True)
-    info = JSONField(null=True)
-
-    oauth_sub = TextField(null=True, unique=True)
-
-    class Meta:
-        database = DB
-
-
-class UserSettings(BaseModel):
-    ui: Optional[dict] = {}
-    model_config = ConfigDict(extra="allow")
-    pass
-
-
-class UserModel(BaseModel):
-    id: str
-    name: str
-    email: str
-    role: str = "pending"
-    profile_image_url: str
-
-    last_active_at: int  # timestamp in epoch
-    updated_at: int  # timestamp in epoch
-    created_at: int  # timestamp in epoch
-
-    api_key: Optional[str] = None
-    settings: Optional[UserSettings] = None
-    info: Optional[dict] = None
-
-    oauth_sub: Optional[str] = None
-
-
-####################
-# Forms
-####################
-
-
-class UserRoleUpdateForm(BaseModel):
-    id: str
-    role: str
-
-
-class UserUpdateForm(BaseModel):
-    name: str
-    email: str
-    profile_image_url: str
-    password: Optional[str] = None
-
-
-class UsersTable:
-    def __init__(self, db):
-        self.db = db
-        self.db.create_tables([User])
-
-    def insert_new_user(
-        self,
-        id: str,
-        name: str,
-        email: str,
-        profile_image_url: str = "/user.png",
-        role: str = "pending",
-        oauth_sub: Optional[str] = None,
-    ) -> Optional[UserModel]:
-        user = UserModel(
-            **{
-                "id": id,
-                "name": name,
-                "email": email,
-                "role": role,
-                "profile_image_url": profile_image_url,
-                "last_active_at": int(time.time()),
-                "created_at": int(time.time()),
-                "updated_at": int(time.time()),
-                "oauth_sub": oauth_sub,
-            }
-        )
-        result = User.create(**user.model_dump())
-        if result:
-            return user
-        else:
-            return None
-
-    def get_user_by_id(self, id: str) -> Optional[UserModel]:
-        try:
-            user = User.get(User.id == id)
-            return UserModel(**model_to_dict(user))
-        except:
-            return None
-
-    def get_user_by_api_key(self, api_key: str) -> Optional[UserModel]:
-        try:
-            user = User.get(User.api_key == api_key)
-            return UserModel(**model_to_dict(user))
-        except:
-            return None
-
-    def get_user_by_email(self, email: str) -> Optional[UserModel]:
-        try:
-            user = User.get(User.email == email)
-            return UserModel(**model_to_dict(user))
-        except:
-            return None
-
-    def get_user_by_oauth_sub(self, sub: str) -> Optional[UserModel]:
-        try:
-            user = User.get(User.oauth_sub == sub)
-            return UserModel(**model_to_dict(user))
-        except:
-            return None
-
-    def get_users(self, skip: int = 0, limit: int = 50) -> List[UserModel]:
-        return [
-            UserModel(**model_to_dict(user))
-            for user in User.select()
-            # .limit(limit).offset(skip)
-        ]
-
-    def get_num_users(self) -> Optional[int]:
-        return User.select().count()
-
-    def get_first_user(self) -> UserModel:
-        try:
-            user = User.select().order_by(User.created_at).first()
-            return UserModel(**model_to_dict(user))
-        except:
-            return None
-
-    def update_user_role_by_id(self, id: str, role: str) -> Optional[UserModel]:
-        try:
-            query = User.update(role=role).where(User.id == id)
-            query.execute()
-
-            user = User.get(User.id == id)
-            return UserModel(**model_to_dict(user))
-        except:
-            return None
-
-    def update_user_profile_image_url_by_id(
-        self, id: str, profile_image_url: str
-    ) -> Optional[UserModel]:
-        try:
-            query = User.update(profile_image_url=profile_image_url).where(
-                User.id == id
-            )
-            query.execute()
-
-            user = User.get(User.id == id)
-            return UserModel(**model_to_dict(user))
-        except:
-            return None
-
-    def update_user_last_active_by_id(self, id: str) -> Optional[UserModel]:
-        try:
-            query = User.update(last_active_at=int(time.time())).where(User.id == id)
-            query.execute()
-
-            user = User.get(User.id == id)
-            return UserModel(**model_to_dict(user))
-        except:
-            return None
-
-    def update_user_oauth_sub_by_id(
-        self, id: str, oauth_sub: str
-    ) -> Optional[UserModel]:
-        try:
-            query = User.update(oauth_sub=oauth_sub).where(User.id == id)
-            query.execute()
-
-            user = User.get(User.id == id)
-            return UserModel(**model_to_dict(user))
-        except:
-            return None
-
-    def update_user_by_id(self, id: str, updated: dict) -> Optional[UserModel]:
-        try:
-            query = User.update(**updated).where(User.id == id)
-            query.execute()
-
-            user = User.get(User.id == id)
-            return UserModel(**model_to_dict(user))
-        except:
-            return None
-
-    def delete_user_by_id(self, id: str) -> bool:
-        try:
-            # Delete User Chats
-            result = Chats.delete_chats_by_user_id(id)
-
-            if result:
-                # Delete User
-                query = User.delete().where(User.id == id)
-                query.execute()  # Remove the rows, return number of rows removed.
-
-                return True
-            else:
-                return False
-        except:
-            return False
-
-    def update_user_api_key_by_id(self, id: str, api_key: str) -> str:
-        try:
-            query = User.update(api_key=api_key).where(User.id == id)
-            result = query.execute()
-
-            return True if result == 1 else False
-        except:
-            return False
-
-    def get_user_api_key_by_id(self, id: str) -> Optional[str]:
-        try:
-            user = User.get(User.id == id)
-            return user.api_key
-        except:
-            return None
-
-
-Users = UsersTable(DB)

backend/apps/webui/utils.py

@@ -1,88 +0,0 @@
-from importlib import util
-import os
-import re
-
-from config import TOOLS_DIR, FUNCTIONS_DIR
-
-
-def extract_frontmatter(file_path):
-    """
-    Extract frontmatter as a dictionary from the specified file path.
-    """
-    frontmatter = {}
-    frontmatter_started = False
-    frontmatter_ended = False
-    frontmatter_pattern = re.compile(r"^\s*([a-z_]+):\s*(.*)\s*$", re.IGNORECASE)
-
-    try:
-        with open(file_path, "r", encoding="utf-8") as file:
-            first_line = file.readline()
-            if first_line.strip() != '"""':
-                # The file doesn't start with triple quotes
-                return {}
-
-            frontmatter_started = True
-
-            for line in file:
-                if '"""' in line:
-                    if frontmatter_started:
-                        frontmatter_ended = True
-                        break
-
-                if frontmatter_started and not frontmatter_ended:
-                    match = frontmatter_pattern.match(line)
-                    if match:
-                        key, value = match.groups()
-                        frontmatter[key.strip()] = value.strip()
-
-    except FileNotFoundError:
-        print(f"Error: The file {file_path} does not exist.")
-        return {}
-    except Exception as e:
-        print(f"An error occurred: {e}")
-        return {}
-
-    return frontmatter
-
-
-def load_toolkit_module_by_id(toolkit_id):
-    toolkit_path = os.path.join(TOOLS_DIR, f"{toolkit_id}.py")
-    spec = util.spec_from_file_location(toolkit_id, toolkit_path)
-    module = util.module_from_spec(spec)
-    frontmatter = extract_frontmatter(toolkit_path)
-
-    try:
-        spec.loader.exec_module(module)
-        print(f"Loaded module: {module.__name__}")
-        if hasattr(module, "Tools"):
-            return module.Tools(), frontmatter
-        else:
-            raise Exception("No Tools class found")
-    except Exception as e:
-        print(f"Error loading module: {toolkit_id}")
-        # Move the file to the error folder
-        os.rename(toolkit_path, f"{toolkit_path}.error")
-        raise e
-
-
-def load_function_module_by_id(function_id):
-    function_path = os.path.join(FUNCTIONS_DIR, f"{function_id}.py")
-
-    spec = util.spec_from_file_location(function_id, function_path)
-    module = util.module_from_spec(spec)
-    frontmatter = extract_frontmatter(function_path)
-
-    try:
-        spec.loader.exec_module(module)
-        print(f"Loaded module: {module.__name__}")
-        if hasattr(module, "Pipe"):
-            return module.Pipe(), "pipe", frontmatter
-        elif hasattr(module, "Filter"):
-            return module.Filter(), "filter", frontmatter
-        else:
-            raise Exception("No Function class found")
-    except Exception as e:
-        print(f"Error loading module: {function_id}")
-        # Move the file to the error folder
-        os.rename(function_path, f"{function_path}.error")
-        raise e

backend/data/config.json

@@ -1,36 +0,0 @@
-{
-	"version": 0,
-	"ui": {
-		"default_locale": "en-US",
-		"prompt_suggestions": [
-			{
-				"title": ["Help me study", "vocabulary for a college entrance exam"],
-				"content": "Help me study vocabulary: write a sentence for me to fill in the blank, and I'll try to pick the correct option."
-			},
-			{
-				"title": ["Give me ideas", "for what to do with my kids' art"],
-				"content": "What are 5 creative things I could do with my kids' art? I don't want to throw them away, but it's also so much clutter."
-			},
-			{
-				"title": ["Tell me a fun fact", "about the Roman Empire"],
-				"content": "Tell me a random fun fact about the Roman Empire"
-			},
-			{
-				"title": ["Show me a code snippet", "of a website's sticky header"],
-				"content": "Show me a code snippet of a website's sticky header in CSS and JavaScript."
-			},
-			{
-				"title": ["Explain options trading", "if I'm familiar with buying and selling stocks"],
-				"content": "Explain options trading in simple terms if I'm familiar with buying and selling stocks."
-			},
-			{
-				"title": ["Overcome procrastination", "give me tips"],
-				"content": "Could you start by asking me about instances when I procrastinate the most and then give me some suggestions to overcome it?"
-			},
-			{
-				"title": ["Grammar check", "rewrite it for better readability "],
-				"content": "Check the following sentence for grammar and clarity: \"[sentence]\". Rewrite it for better readability while maintaining its original meaning."
-			}
-		]
-	}
-}

backend/data/litellm/config.yaml

@@ -1,4 +0,0 @@
-general_settings: {}
-litellm_settings: {}
-model_list: []
-router_settings: {}

backend/data/readme.txt

@@ -1 +1 @@
-dir for backend files (db, documents, etc.)
+docker dir for backend files (db, documents, etc.)

backend/dev.sh

@@ -1,2 +1,2 @@
 PORT="${PORT:-8080}"
-uvicorn main:app --port $PORT --host 0.0.0.0 --forwarded-allow-ips '*' --reload
+uvicorn open_webui.main:app --port $PORT --host 0.0.0.0 --forwarded-allow-ips '*' --reload

backend/open_webui/__init__.py

@@ -9,8 +9,6 @@ import uvicorn
 app = typer.Typer()
 
 KEY_FILE = Path.cwd() / ".webui_secret_key"
-if (frontend_build_dir := Path(__file__).parent / "frontend").exists():
-    os.environ["FRONTEND_BUILD_DIR"] = str(frontend_build_dir)
 
 
 @app.command()
@@ -18,6 +16,7 @@ def serve(
     host: str = "0.0.0.0",
     port: int = 8080,
 ):
+    os.environ["FROM_INIT_PY"] = "true"
     if os.getenv("WEBUI_SECRET_KEY") is None:
         typer.echo(
             "Loading WEBUI_SECRET_KEY from file, not provided as an environment variable."
@@ -40,9 +39,23 @@ def serve(
                 "/usr/local/lib/python3.11/site-packages/nvidia/cudnn/lib",
             ]
         )
-    import main  # we need set environment variables before importing main
+        try:
+            import torch
 
-    uvicorn.run(main.app, host=host, port=port, forwarded_allow_ips="*")
+            assert torch.cuda.is_available(), "CUDA not available"
+            typer.echo("CUDA seems to be working")
+        except Exception as e:
+            typer.echo(
+                "Error when testing CUDA but USE_CUDA_DOCKER is true. "
+                "Resetting USE_CUDA_DOCKER to false and removing "
+                f"LD_LIBRARY_PATH modifications: {e}"
+            )
+            os.environ["USE_CUDA_DOCKER"] = "false"
+            os.environ["LD_LIBRARY_PATH"] = ":".join(LD_LIBRARY_PATH)
+
+    import open_webui.main  # we need set environment variables before importing main
+
+    uvicorn.run(open_webui.main.app, host=host, port=port, forwarded_allow_ips="*")
 
 
 @app.command()
@@ -52,7 +65,11 @@ def dev(
     reload: bool = True,
 ):
     uvicorn.run(
-        "main:app", host=host, port=port, reload=reload, forwarded_allow_ips="*"
+        "open_webui.main:app",
+        host=host,
+        port=port,
+        reload=reload,
+        forwarded_allow_ips="*",
     )
 
 

backend/open_webui/alembic.ini

@@ -0,0 +1,114 @@
+# A generic, single database configuration.
+
+[alembic]
+# path to migration scripts
+script_location = migrations
+
+# template used to generate migration file names; The default value is %%(rev)s_%%(slug)s
+# Uncomment the line below if you want the files to be prepended with date and time
+# file_template = %%(year)d_%%(month).2d_%%(day).2d_%%(hour).2d%%(minute).2d-%%(rev)s_%%(slug)s
+
+# sys.path path, will be prepended to sys.path if present.
+# defaults to the current working directory.
+prepend_sys_path = .
+
+# timezone to use when rendering the date within the migration file
+# as well as the filename.
+# If specified, requires the python>=3.9 or backports.zoneinfo library.
+# Any required deps can installed by adding `alembic[tz]` to the pip requirements
+# string value is passed to ZoneInfo()
+# leave blank for localtime
+# timezone =
+
+# max length of characters to apply to the
+# "slug" field
+# truncate_slug_length = 40
+
+# set to 'true' to run the environment during
+# the 'revision' command, regardless of autogenerate
+# revision_environment = false
+
+# set to 'true' to allow .pyc and .pyo files without
+# a source .py file to be detected as revisions in the
+# versions/ directory
+# sourceless = false
+
+# version location specification; This defaults
+# to migrations/versions.  When using multiple version
+# directories, initial revisions must be specified with --version-path.
+# The path separator used here should be the separator specified by "version_path_separator" below.
+# version_locations = %(here)s/bar:%(here)s/bat:migrations/versions
+
+# version path separator; As mentioned above, this is the character used to split
+# version_locations. The default within new alembic.ini files is "os", which uses os.pathsep.
+# If this key is omitted entirely, it falls back to the legacy behavior of splitting on spaces and/or commas.
+# Valid values for version_path_separator are:
+#
+# version_path_separator = :
+# version_path_separator = ;
+# version_path_separator = space
+version_path_separator = os  # Use os.pathsep. Default configuration used for new projects.
+
+# set to 'true' to search source files recursively
+# in each "version_locations" directory
+# new in Alembic version 1.10
+# recursive_version_locations = false
+
+# the output encoding used when revision files
+# are written from script.py.mako
+# output_encoding = utf-8
+
+# sqlalchemy.url = REPLACE_WITH_DATABASE_URL
+
+
+[post_write_hooks]
+# post_write_hooks defines scripts or Python functions that are run
+# on newly generated revision scripts.  See the documentation for further
+# detail and examples
+
+# format using "black" - use the console_scripts runner, against the "black" entrypoint
+# hooks = black
+# black.type = console_scripts
+# black.entrypoint = black
+# black.options = -l 79 REVISION_SCRIPT_FILENAME
+
+# lint with attempts to fix using "ruff" - use the exec runner, execute a binary
+# hooks = ruff
+# ruff.type = exec
+# ruff.executable = %(here)s/.venv/bin/ruff
+# ruff.options = --fix REVISION_SCRIPT_FILENAME
+
+# Logging configuration
+[loggers]
+keys = root,sqlalchemy,alembic
+
+[handlers]
+keys = console
+
+[formatters]
+keys = generic
+
+[logger_root]
+level = WARN
+handlers = console
+qualname =
+
+[logger_sqlalchemy]
+level = WARN
+handlers =
+qualname = sqlalchemy.engine
+
+[logger_alembic]
+level = INFO
+handlers =
+qualname = alembic
+
+[handler_console]
+class = StreamHandler
+args = (sys.stderr,)
+level = NOTSET
+formatter = generic
+
+[formatter_generic]
+format = %(levelname)-5.5s [%(name)s] %(message)s
+datefmt = %H:%M:%S

backend/open_webui/apps/audio/main.py

@@ -0,0 +1,583 @@
+import hashlib
+import json
+import logging
+import os
+import uuid
+from functools import lru_cache
+from pathlib import Path
+
+import requests
+from open_webui.config import (
+    AUDIO_STT_ENGINE,
+    AUDIO_STT_MODEL,
+    AUDIO_STT_OPENAI_API_BASE_URL,
+    AUDIO_STT_OPENAI_API_KEY,
+    AUDIO_TTS_API_KEY,
+    AUDIO_TTS_ENGINE,
+    AUDIO_TTS_MODEL,
+    AUDIO_TTS_OPENAI_API_BASE_URL,
+    AUDIO_TTS_OPENAI_API_KEY,
+    AUDIO_TTS_SPLIT_ON,
+    AUDIO_TTS_VOICE,
+    AUDIO_TTS_AZURE_SPEECH_REGION,
+    AUDIO_TTS_AZURE_SPEECH_OUTPUT_FORMAT,
+    CACHE_DIR,
+    CORS_ALLOW_ORIGIN,
+    WHISPER_MODEL,
+    WHISPER_MODEL_AUTO_UPDATE,
+    WHISPER_MODEL_DIR,
+    AppConfig,
+)
+
+from open_webui.constants import ERROR_MESSAGES
+from open_webui.env import SRC_LOG_LEVELS, DEVICE_TYPE
+from fastapi import Depends, FastAPI, File, HTTPException, Request, UploadFile, status
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import FileResponse
+from pydantic import BaseModel
+from open_webui.utils.utils import get_admin_user, get_current_user, get_verified_user
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["AUDIO"])
+
+app = FastAPI()
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=CORS_ALLOW_ORIGIN,
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+app.state.config = AppConfig()
+
+app.state.config.STT_OPENAI_API_BASE_URL = AUDIO_STT_OPENAI_API_BASE_URL
+app.state.config.STT_OPENAI_API_KEY = AUDIO_STT_OPENAI_API_KEY
+app.state.config.STT_ENGINE = AUDIO_STT_ENGINE
+app.state.config.STT_MODEL = AUDIO_STT_MODEL
+
+app.state.config.TTS_OPENAI_API_BASE_URL = AUDIO_TTS_OPENAI_API_BASE_URL
+app.state.config.TTS_OPENAI_API_KEY = AUDIO_TTS_OPENAI_API_KEY
+app.state.config.TTS_ENGINE = AUDIO_TTS_ENGINE
+app.state.config.TTS_MODEL = AUDIO_TTS_MODEL
+app.state.config.TTS_VOICE = AUDIO_TTS_VOICE
+app.state.config.TTS_API_KEY = AUDIO_TTS_API_KEY
+app.state.config.TTS_SPLIT_ON = AUDIO_TTS_SPLIT_ON
+
+app.state.config.TTS_AZURE_SPEECH_REGION = AUDIO_TTS_AZURE_SPEECH_REGION
+app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT = AUDIO_TTS_AZURE_SPEECH_OUTPUT_FORMAT
+
+# setting device type for whisper model
+whisper_device_type = DEVICE_TYPE if DEVICE_TYPE and DEVICE_TYPE == "cuda" else "cpu"
+log.info(f"whisper_device_type: {whisper_device_type}")
+
+SPEECH_CACHE_DIR = Path(CACHE_DIR).joinpath("./audio/speech/")
+SPEECH_CACHE_DIR.mkdir(parents=True, exist_ok=True)
+
+
+class TTSConfigForm(BaseModel):
+    OPENAI_API_BASE_URL: str
+    OPENAI_API_KEY: str
+    API_KEY: str
+    ENGINE: str
+    MODEL: str
+    VOICE: str
+    SPLIT_ON: str
+    AZURE_SPEECH_REGION: str
+    AZURE_SPEECH_OUTPUT_FORMAT: str
+
+
+class STTConfigForm(BaseModel):
+    OPENAI_API_BASE_URL: str
+    OPENAI_API_KEY: str
+    ENGINE: str
+    MODEL: str
+
+
+class AudioConfigUpdateForm(BaseModel):
+    tts: TTSConfigForm
+    stt: STTConfigForm
+
+
+from pydub import AudioSegment
+from pydub.utils import mediainfo
+
+
+def is_mp4_audio(file_path):
+    """Check if the given file is an MP4 audio file."""
+    if not os.path.isfile(file_path):
+        print(f"File not found: {file_path}")
+        return False
+
+    info = mediainfo(file_path)
+    if (
+        info.get("codec_name") == "aac"
+        and info.get("codec_type") == "audio"
+        and info.get("codec_tag_string") == "mp4a"
+    ):
+        return True
+    return False
+
+
+def convert_mp4_to_wav(file_path, output_path):
+    """Convert MP4 audio file to WAV format."""
+    audio = AudioSegment.from_file(file_path, format="mp4")
+    audio.export(output_path, format="wav")
+    print(f"Converted {file_path} to {output_path}")
+
+
+@app.get("/config")
+async def get_audio_config(user=Depends(get_admin_user)):
+    return {
+        "tts": {
+            "OPENAI_API_BASE_URL": app.state.config.TTS_OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": app.state.config.TTS_OPENAI_API_KEY,
+            "API_KEY": app.state.config.TTS_API_KEY,
+            "ENGINE": app.state.config.TTS_ENGINE,
+            "MODEL": app.state.config.TTS_MODEL,
+            "VOICE": app.state.config.TTS_VOICE,
+            "SPLIT_ON": app.state.config.TTS_SPLIT_ON,
+            "AZURE_SPEECH_REGION": app.state.config.TTS_AZURE_SPEECH_REGION,
+            "AZURE_SPEECH_OUTPUT_FORMAT": app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT,
+        },
+        "stt": {
+            "OPENAI_API_BASE_URL": app.state.config.STT_OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": app.state.config.STT_OPENAI_API_KEY,
+            "ENGINE": app.state.config.STT_ENGINE,
+            "MODEL": app.state.config.STT_MODEL,
+        },
+    }
+
+
+@app.post("/config/update")
+async def update_audio_config(
+    form_data: AudioConfigUpdateForm, user=Depends(get_admin_user)
+):
+    app.state.config.TTS_OPENAI_API_BASE_URL = form_data.tts.OPENAI_API_BASE_URL
+    app.state.config.TTS_OPENAI_API_KEY = form_data.tts.OPENAI_API_KEY
+    app.state.config.TTS_API_KEY = form_data.tts.API_KEY
+    app.state.config.TTS_ENGINE = form_data.tts.ENGINE
+    app.state.config.TTS_MODEL = form_data.tts.MODEL
+    app.state.config.TTS_VOICE = form_data.tts.VOICE
+    app.state.config.TTS_SPLIT_ON = form_data.tts.SPLIT_ON
+    app.state.config.TTS_AZURE_SPEECH_REGION = form_data.tts.AZURE_SPEECH_REGION
+    app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT = (
+        form_data.tts.AZURE_SPEECH_OUTPUT_FORMAT
+    )
+
+    app.state.config.STT_OPENAI_API_BASE_URL = form_data.stt.OPENAI_API_BASE_URL
+    app.state.config.STT_OPENAI_API_KEY = form_data.stt.OPENAI_API_KEY
+    app.state.config.STT_ENGINE = form_data.stt.ENGINE
+    app.state.config.STT_MODEL = form_data.stt.MODEL
+
+    return {
+        "tts": {
+            "OPENAI_API_BASE_URL": app.state.config.TTS_OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": app.state.config.TTS_OPENAI_API_KEY,
+            "API_KEY": app.state.config.TTS_API_KEY,
+            "ENGINE": app.state.config.TTS_ENGINE,
+            "MODEL": app.state.config.TTS_MODEL,
+            "VOICE": app.state.config.TTS_VOICE,
+            "SPLIT_ON": app.state.config.TTS_SPLIT_ON,
+            "AZURE_SPEECH_REGION": app.state.config.TTS_AZURE_SPEECH_REGION,
+            "AZURE_SPEECH_OUTPUT_FORMAT": app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT,
+        },
+        "stt": {
+            "OPENAI_API_BASE_URL": app.state.config.STT_OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": app.state.config.STT_OPENAI_API_KEY,
+            "ENGINE": app.state.config.STT_ENGINE,
+            "MODEL": app.state.config.STT_MODEL,
+        },
+    }
+
+
+@app.post("/speech")
+async def speech(request: Request, user=Depends(get_verified_user)):
+    body = await request.body()
+    name = hashlib.sha256(body).hexdigest()
+
+    file_path = SPEECH_CACHE_DIR.joinpath(f"{name}.mp3")
+    file_body_path = SPEECH_CACHE_DIR.joinpath(f"{name}.json")
+
+    # Check if the file already exists in the cache
+    if file_path.is_file():
+        return FileResponse(file_path)
+
+    if app.state.config.TTS_ENGINE == "openai":
+        headers = {}
+        headers["Authorization"] = f"Bearer {app.state.config.TTS_OPENAI_API_KEY}"
+        headers["Content-Type"] = "application/json"
+
+        try:
+            body = body.decode("utf-8")
+            body = json.loads(body)
+            body["model"] = app.state.config.TTS_MODEL
+            body = json.dumps(body).encode("utf-8")
+        except Exception:
+            pass
+
+        r = None
+        try:
+            r = requests.post(
+                url=f"{app.state.config.TTS_OPENAI_API_BASE_URL}/audio/speech",
+                data=body,
+                headers=headers,
+                stream=True,
+            )
+
+            r.raise_for_status()
+
+            # Save the streaming content to a file
+            with open(file_path, "wb") as f:
+                for chunk in r.iter_content(chunk_size=8192):
+                    f.write(chunk)
+
+            with open(file_body_path, "w") as f:
+                json.dump(json.loads(body.decode("utf-8")), f)
+
+            # Return the saved file
+            return FileResponse(file_path)
+
+        except Exception as e:
+            log.exception(e)
+            error_detail = "Open WebUI: Server Connection Error"
+            if r is not None:
+                try:
+                    res = r.json()
+                    if "error" in res:
+                        error_detail = f"External: {res['error']['message']}"
+                except Exception:
+                    error_detail = f"External: {e}"
+
+            raise HTTPException(
+                status_code=r.status_code if r != None else 500,
+                detail=error_detail,
+            )
+
+    elif app.state.config.TTS_ENGINE == "elevenlabs":
+        payload = None
+        try:
+            payload = json.loads(body.decode("utf-8"))
+        except Exception as e:
+            log.exception(e)
+            raise HTTPException(status_code=400, detail="Invalid JSON payload")
+
+        voice_id = payload.get("voice", "")
+
+        if voice_id not in get_available_voices():
+            raise HTTPException(
+                status_code=400,
+                detail="Invalid voice id",
+            )
+
+        url = f"https://api.elevenlabs.io/v1/text-to-speech/{voice_id}"
+
+        headers = {
+            "Accept": "audio/mpeg",
+            "Content-Type": "application/json",
+            "xi-api-key": app.state.config.TTS_API_KEY,
+        }
+
+        data = {
+            "text": payload["input"],
+            "model_id": app.state.config.TTS_MODEL,
+            "voice_settings": {"stability": 0.5, "similarity_boost": 0.5},
+        }
+
+        try:
+            r = requests.post(url, json=data, headers=headers)
+
+            r.raise_for_status()
+
+            # Save the streaming content to a file
+            with open(file_path, "wb") as f:
+                for chunk in r.iter_content(chunk_size=8192):
+                    f.write(chunk)
+
+            with open(file_body_path, "w") as f:
+                json.dump(json.loads(body.decode("utf-8")), f)
+
+            # Return the saved file
+            return FileResponse(file_path)
+
+        except Exception as e:
+            log.exception(e)
+            error_detail = "Open WebUI: Server Connection Error"
+            if r is not None:
+                try:
+                    res = r.json()
+                    if "error" in res:
+                        error_detail = f"External: {res['error']['message']}"
+                except Exception:
+                    error_detail = f"External: {e}"
+
+            raise HTTPException(
+                status_code=r.status_code if r != None else 500,
+                detail=error_detail,
+            )
+
+    elif app.state.config.TTS_ENGINE == "azure":
+        payload = None
+        try:
+            payload = json.loads(body.decode("utf-8"))
+        except Exception as e:
+            log.exception(e)
+            raise HTTPException(status_code=400, detail="Invalid JSON payload")
+
+        region = app.state.config.TTS_AZURE_SPEECH_REGION
+        language = app.state.config.TTS_VOICE
+        locale = "-".join(app.state.config.TTS_VOICE.split("-")[:1])
+        output_format = app.state.config.TTS_AZURE_SPEECH_OUTPUT_FORMAT
+        url = f"https://{region}.tts.speech.microsoft.com/cognitiveservices/v1"
+
+        headers = {
+            "Ocp-Apim-Subscription-Key": app.state.config.TTS_API_KEY,
+            "Content-Type": "application/ssml+xml",
+            "X-Microsoft-OutputFormat": output_format,
+        }
+
+        data = f"""<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="{locale}">
+                <voice name="{language}">{payload["input"]}</voice>
+            </speak>"""
+
+        response = requests.post(url, headers=headers, data=data)
+
+        if response.status_code == 200:
+            with open(file_path, "wb") as f:
+                f.write(response.content)
+            return FileResponse(file_path)
+        else:
+            log.error(f"Error synthesizing speech - {response.reason}")
+            raise HTTPException(
+                status_code=500, detail=f"Error synthesizing speech - {response.reason}"
+            )
+
+
+@app.post("/transcriptions")
+def transcribe(
+    file: UploadFile = File(...),
+    user=Depends(get_current_user),
+):
+    log.info(f"file.content_type: {file.content_type}")
+
+    if file.content_type not in ["audio/mpeg", "audio/wav", "audio/ogg", "audio/x-m4a"]:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.FILE_NOT_SUPPORTED,
+        )
+
+    try:
+        ext = file.filename.split(".")[-1]
+
+        id = uuid.uuid4()
+        filename = f"{id}.{ext}"
+
+        file_dir = f"{CACHE_DIR}/audio/transcriptions"
+        os.makedirs(file_dir, exist_ok=True)
+        file_path = f"{file_dir}/{filename}"
+
+        print(filename)
+
+        contents = file.file.read()
+        with open(file_path, "wb") as f:
+            f.write(contents)
+            f.close()
+
+        if app.state.config.STT_ENGINE == "":
+            from faster_whisper import WhisperModel
+
+            whisper_kwargs = {
+                "model_size_or_path": WHISPER_MODEL,
+                "device": whisper_device_type,
+                "compute_type": "int8",
+                "download_root": WHISPER_MODEL_DIR,
+                "local_files_only": not WHISPER_MODEL_AUTO_UPDATE,
+            }
+
+            log.debug(f"whisper_kwargs: {whisper_kwargs}")
+
+            try:
+                model = WhisperModel(**whisper_kwargs)
+            except Exception:
+                log.warning(
+                    "WhisperModel initialization failed, attempting download with local_files_only=False"
+                )
+                whisper_kwargs["local_files_only"] = False
+                model = WhisperModel(**whisper_kwargs)
+
+            segments, info = model.transcribe(file_path, beam_size=5)
+            log.info(
+                "Detected language '%s' with probability %f"
+                % (info.language, info.language_probability)
+            )
+
+            transcript = "".join([segment.text for segment in list(segments)])
+
+            data = {"text": transcript.strip()}
+
+            # save the transcript to a json file
+            transcript_file = f"{file_dir}/{id}.json"
+            with open(transcript_file, "w") as f:
+                json.dump(data, f)
+
+            print(data)
+
+            return data
+
+        elif app.state.config.STT_ENGINE == "openai":
+            if is_mp4_audio(file_path):
+                print("is_mp4_audio")
+                os.rename(file_path, file_path.replace(".wav", ".mp4"))
+                # Convert MP4 audio file to WAV format
+                convert_mp4_to_wav(file_path.replace(".wav", ".mp4"), file_path)
+
+            headers = {"Authorization": f"Bearer {app.state.config.STT_OPENAI_API_KEY}"}
+
+            files = {"file": (filename, open(file_path, "rb"))}
+            data = {"model": app.state.config.STT_MODEL}
+
+            print(files, data)
+
+            r = None
+            try:
+                r = requests.post(
+                    url=f"{app.state.config.STT_OPENAI_API_BASE_URL}/audio/transcriptions",
+                    headers=headers,
+                    files=files,
+                    data=data,
+                )
+
+                r.raise_for_status()
+
+                data = r.json()
+
+                # save the transcript to a json file
+                transcript_file = f"{file_dir}/{id}.json"
+                with open(transcript_file, "w") as f:
+                    json.dump(data, f)
+
+                print(data)
+                return data
+            except Exception as e:
+                log.exception(e)
+                error_detail = "Open WebUI: Server Connection Error"
+                if r is not None:
+                    try:
+                        res = r.json()
+                        if "error" in res:
+                            error_detail = f"External: {res['error']['message']}"
+                    except Exception:
+                        error_detail = f"External: {e}"
+
+                raise HTTPException(
+                    status_code=r.status_code if r != None else 500,
+                    detail=error_detail,
+                )
+
+    except Exception as e:
+        log.exception(e)
+
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail=ERROR_MESSAGES.DEFAULT(e),
+        )
+
+
+def get_available_models() -> list[dict]:
+    if app.state.config.TTS_ENGINE == "openai":
+        return [{"id": "tts-1"}, {"id": "tts-1-hd"}]
+    elif app.state.config.TTS_ENGINE == "elevenlabs":
+        headers = {
+            "xi-api-key": app.state.config.TTS_API_KEY,
+            "Content-Type": "application/json",
+        }
+
+        try:
+            response = requests.get(
+                "https://api.elevenlabs.io/v1/models", headers=headers, timeout=5
+            )
+            response.raise_for_status()
+            models = response.json()
+            return [
+                {"name": model["name"], "id": model["model_id"]} for model in models
+            ]
+        except requests.RequestException as e:
+            log.error(f"Error fetching voices: {str(e)}")
+    return []
+
+
+@app.get("/models")
+async def get_models(user=Depends(get_verified_user)):
+    return {"models": get_available_models()}
+
+
+def get_available_voices() -> dict:
+    """Returns {voice_id: voice_name} dict"""
+    ret = {}
+    if app.state.config.TTS_ENGINE == "openai":
+        ret = {
+            "alloy": "alloy",
+            "echo": "echo",
+            "fable": "fable",
+            "onyx": "onyx",
+            "nova": "nova",
+            "shimmer": "shimmer",
+        }
+    elif app.state.config.TTS_ENGINE == "elevenlabs":
+        try:
+            ret = get_elevenlabs_voices()
+        except Exception:
+            # Avoided @lru_cache with exception
+            pass
+    elif app.state.config.TTS_ENGINE == "azure":
+        try:
+            region = app.state.config.TTS_AZURE_SPEECH_REGION
+            url = f"https://{region}.tts.speech.microsoft.com/cognitiveservices/voices/list"
+            headers = {"Ocp-Apim-Subscription-Key": app.state.config.TTS_API_KEY}
+
+            response = requests.get(url, headers=headers)
+            response.raise_for_status()
+            voices = response.json()
+            for voice in voices:
+                ret[voice["ShortName"]] = (
+                    f"{voice['DisplayName']} ({voice['ShortName']})"
+                )
+        except requests.RequestException as e:
+            log.error(f"Error fetching voices: {str(e)}")
+
+    return ret
+
+
+@lru_cache
+def get_elevenlabs_voices() -> dict:
+    """
+    Note, set the following in your .env file to use Elevenlabs:
+    AUDIO_TTS_ENGINE=elevenlabs
+    AUDIO_TTS_API_KEY=sk_...  # Your Elevenlabs API key
+    AUDIO_TTS_VOICE=EXAVITQu4vr4xnSDxMaL  # From https://api.elevenlabs.io/v1/voices
+    AUDIO_TTS_MODEL=eleven_multilingual_v2
+    """
+    headers = {
+        "xi-api-key": app.state.config.TTS_API_KEY,
+        "Content-Type": "application/json",
+    }
+    try:
+        # TODO: Add retries
+        response = requests.get("https://api.elevenlabs.io/v1/voices", headers=headers)
+        response.raise_for_status()
+        voices_data = response.json()
+
+        voices = {}
+        for voice in voices_data.get("voices", []):
+            voices[voice["voice_id"]] = voice["name"]
+    except requests.RequestException as e:
+        # Avoid @lru_cache with exception
+        log.error(f"Error fetching voices: {str(e)}")
+        raise RuntimeError(f"Error fetching voices: {str(e)}")
+
+    return voices
+
+
+@app.get("/voices")
+async def get_voices(user=Depends(get_verified_user)):
+    return {"voices": [{"id": k, "name": v} for k, v in get_available_voices().items()]}

backend/open_webui/apps/images/main.py

@@ -1,55 +1,45 @@
-import re
-import requests
-import base64
-from fastapi import (
-    FastAPI,
-    Request,
-    Depends,
-    HTTPException,
-    status,
-    UploadFile,
-    File,
-    Form,
-)
-from fastapi.middleware.cors import CORSMiddleware
-from faster_whisper import WhisperModel
-
-from constants import ERROR_MESSAGES
-from utils.utils import (
-    get_verified_user,
-    get_admin_user,
-)
-
-from apps.images.utils.comfyui import ImageGenerationPayload, comfyui_generate_image
-from utils.misc import calculate_sha256
-from typing import Optional
-from pydantic import BaseModel
-from pathlib import Path
-import mimetypes
-import uuid
+import asyncio
 import base64
 import json
 import logging
+import mimetypes
+import re
+import uuid
+from pathlib import Path
+from typing import Optional
 
-from config import (
-    SRC_LOG_LEVELS,
-    CACHE_DIR,
-    IMAGE_GENERATION_ENGINE,
-    ENABLE_IMAGE_GENERATION,
-    AUTOMATIC1111_BASE_URL,
+import requests
+from open_webui.apps.images.utils.comfyui import (
+    ComfyUIGenerateImageForm,
+    ComfyUIWorkflow,
+    comfyui_generate_image,
+)
+from open_webui.config import (
     AUTOMATIC1111_API_AUTH,
+    AUTOMATIC1111_BASE_URL,
+    AUTOMATIC1111_CFG_SCALE,
+    AUTOMATIC1111_SAMPLER,
+    AUTOMATIC1111_SCHEDULER,
+    CACHE_DIR,
     COMFYUI_BASE_URL,
-    COMFYUI_CFG_SCALE,
-    COMFYUI_SAMPLER,
-    COMFYUI_SCHEDULER,
-    COMFYUI_SD3,
-    IMAGES_OPENAI_API_BASE_URL,
-    IMAGES_OPENAI_API_KEY,
+    COMFYUI_WORKFLOW,
+    COMFYUI_WORKFLOW_NODES,
+    CORS_ALLOW_ORIGIN,
+    ENABLE_IMAGE_GENERATION,
+    IMAGE_GENERATION_ENGINE,
     IMAGE_GENERATION_MODEL,
     IMAGE_SIZE,
     IMAGE_STEPS,
+    IMAGES_OPENAI_API_BASE_URL,
+    IMAGES_OPENAI_API_KEY,
     AppConfig,
 )
+from open_webui.constants import ERROR_MESSAGES
+from open_webui.env import SRC_LOG_LEVELS
+from fastapi import Depends, FastAPI, HTTPException, Request
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+from open_webui.utils.utils import get_admin_user, get_verified_user
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["IMAGES"])
@@ -60,7 +50,7 @@ IMAGE_CACHE_DIR.mkdir(parents=True, exist_ok=True)
 app = FastAPI()
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=["*"],
+    allow_origins=CORS_ALLOW_ORIGIN,
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],
@@ -78,18 +68,127 @@ app.state.config.MODEL = IMAGE_GENERATION_MODEL
 
 app.state.config.AUTOMATIC1111_BASE_URL = AUTOMATIC1111_BASE_URL
 app.state.config.AUTOMATIC1111_API_AUTH = AUTOMATIC1111_API_AUTH
+app.state.config.AUTOMATIC1111_CFG_SCALE = AUTOMATIC1111_CFG_SCALE
+app.state.config.AUTOMATIC1111_SAMPLER = AUTOMATIC1111_SAMPLER
+app.state.config.AUTOMATIC1111_SCHEDULER = AUTOMATIC1111_SCHEDULER
 app.state.config.COMFYUI_BASE_URL = COMFYUI_BASE_URL
+app.state.config.COMFYUI_WORKFLOW = COMFYUI_WORKFLOW
+app.state.config.COMFYUI_WORKFLOW_NODES = COMFYUI_WORKFLOW_NODES
 
 app.state.config.IMAGE_SIZE = IMAGE_SIZE
 app.state.config.IMAGE_STEPS = IMAGE_STEPS
-app.state.config.COMFYUI_CFG_SCALE = COMFYUI_CFG_SCALE
-app.state.config.COMFYUI_SAMPLER = COMFYUI_SAMPLER
-app.state.config.COMFYUI_SCHEDULER = COMFYUI_SCHEDULER
-app.state.config.COMFYUI_SD3 = COMFYUI_SD3
+
+
+@app.get("/config")
+async def get_config(request: Request, user=Depends(get_admin_user)):
+    return {
+        "enabled": app.state.config.ENABLED,
+        "engine": app.state.config.ENGINE,
+        "openai": {
+            "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY,
+        },
+        "automatic1111": {
+            "AUTOMATIC1111_BASE_URL": app.state.config.AUTOMATIC1111_BASE_URL,
+            "AUTOMATIC1111_API_AUTH": app.state.config.AUTOMATIC1111_API_AUTH,
+            "AUTOMATIC1111_CFG_SCALE": app.state.config.AUTOMATIC1111_CFG_SCALE,
+            "AUTOMATIC1111_SAMPLER": app.state.config.AUTOMATIC1111_SAMPLER,
+            "AUTOMATIC1111_SCHEDULER": app.state.config.AUTOMATIC1111_SCHEDULER,
+        },
+        "comfyui": {
+            "COMFYUI_BASE_URL": app.state.config.COMFYUI_BASE_URL,
+            "COMFYUI_WORKFLOW": app.state.config.COMFYUI_WORKFLOW,
+            "COMFYUI_WORKFLOW_NODES": app.state.config.COMFYUI_WORKFLOW_NODES,
+        },
+    }
+
+
+class OpenAIConfigForm(BaseModel):
+    OPENAI_API_BASE_URL: str
+    OPENAI_API_KEY: str
+
+
+class Automatic1111ConfigForm(BaseModel):
+    AUTOMATIC1111_BASE_URL: str
+    AUTOMATIC1111_API_AUTH: str
+    AUTOMATIC1111_CFG_SCALE: Optional[str]
+    AUTOMATIC1111_SAMPLER: Optional[str]
+    AUTOMATIC1111_SCHEDULER: Optional[str]
+
+
+class ComfyUIConfigForm(BaseModel):
+    COMFYUI_BASE_URL: str
+    COMFYUI_WORKFLOW: str
+    COMFYUI_WORKFLOW_NODES: list[dict]
+
+
+class ConfigForm(BaseModel):
+    enabled: bool
+    engine: str
+    openai: OpenAIConfigForm
+    automatic1111: Automatic1111ConfigForm
+    comfyui: ComfyUIConfigForm
+
+
+@app.post("/config/update")
+async def update_config(form_data: ConfigForm, user=Depends(get_admin_user)):
+    app.state.config.ENGINE = form_data.engine
+    app.state.config.ENABLED = form_data.enabled
+
+    app.state.config.OPENAI_API_BASE_URL = form_data.openai.OPENAI_API_BASE_URL
+    app.state.config.OPENAI_API_KEY = form_data.openai.OPENAI_API_KEY
+
+    app.state.config.AUTOMATIC1111_BASE_URL = (
+        form_data.automatic1111.AUTOMATIC1111_BASE_URL
+    )
+    app.state.config.AUTOMATIC1111_API_AUTH = (
+        form_data.automatic1111.AUTOMATIC1111_API_AUTH
+    )
+
+    app.state.config.AUTOMATIC1111_CFG_SCALE = (
+        float(form_data.automatic1111.AUTOMATIC1111_CFG_SCALE)
+        if form_data.automatic1111.AUTOMATIC1111_CFG_SCALE
+        else None
+    )
+    app.state.config.AUTOMATIC1111_SAMPLER = (
+        form_data.automatic1111.AUTOMATIC1111_SAMPLER
+        if form_data.automatic1111.AUTOMATIC1111_SAMPLER
+        else None
+    )
+    app.state.config.AUTOMATIC1111_SCHEDULER = (
+        form_data.automatic1111.AUTOMATIC1111_SCHEDULER
+        if form_data.automatic1111.AUTOMATIC1111_SCHEDULER
+        else None
+    )
+
+    app.state.config.COMFYUI_BASE_URL = form_data.comfyui.COMFYUI_BASE_URL.strip("/")
+    app.state.config.COMFYUI_WORKFLOW = form_data.comfyui.COMFYUI_WORKFLOW
+    app.state.config.COMFYUI_WORKFLOW_NODES = form_data.comfyui.COMFYUI_WORKFLOW_NODES
+
+    return {
+        "enabled": app.state.config.ENABLED,
+        "engine": app.state.config.ENGINE,
+        "openai": {
+            "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL,
+            "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY,
+        },
+        "automatic1111": {
+            "AUTOMATIC1111_BASE_URL": app.state.config.AUTOMATIC1111_BASE_URL,
+            "AUTOMATIC1111_API_AUTH": app.state.config.AUTOMATIC1111_API_AUTH,
+            "AUTOMATIC1111_CFG_SCALE": app.state.config.AUTOMATIC1111_CFG_SCALE,
+            "AUTOMATIC1111_SAMPLER": app.state.config.AUTOMATIC1111_SAMPLER,
+            "AUTOMATIC1111_SCHEDULER": app.state.config.AUTOMATIC1111_SCHEDULER,
+        },
+        "comfyui": {
+            "COMFYUI_BASE_URL": app.state.config.COMFYUI_BASE_URL,
+            "COMFYUI_WORKFLOW": app.state.config.COMFYUI_WORKFLOW,
+            "COMFYUI_WORKFLOW_NODES": app.state.config.COMFYUI_WORKFLOW_NODES,
+        },
+    }
 
 
 def get_automatic1111_api_auth():
-    if app.state.config.AUTOMATIC1111_API_AUTH == None:
+    if app.state.config.AUTOMATIC1111_API_AUTH is None:
         return ""
     else:
         auth1111_byte_string = app.state.config.AUTOMATIC1111_API_AUTH.encode("utf-8")
@@ -98,164 +197,112 @@ def get_automatic1111_api_auth():
         return f"Basic {auth1111_base64_encoded_string}"
 
 
-@app.get("/config")
-async def get_config(request: Request, user=Depends(get_admin_user)):
-    return {
-        "engine": app.state.config.ENGINE,
-        "enabled": app.state.config.ENABLED,
-    }
-
-
-class ConfigUpdateForm(BaseModel):
-    engine: str
-    enabled: bool
-
-
-@app.post("/config/update")
-async def update_config(form_data: ConfigUpdateForm, user=Depends(get_admin_user)):
-    app.state.config.ENGINE = form_data.engine
-    app.state.config.ENABLED = form_data.enabled
-    return {
-        "engine": app.state.config.ENGINE,
-        "enabled": app.state.config.ENABLED,
-    }
-
-
-class EngineUrlUpdateForm(BaseModel):
-    AUTOMATIC1111_BASE_URL: Optional[str] = None
-    AUTOMATIC1111_API_AUTH: Optional[str] = None
-    COMFYUI_BASE_URL: Optional[str] = None
-
-
-@app.get("/url")
-async def get_engine_url(user=Depends(get_admin_user)):
-    return {
-        "AUTOMATIC1111_BASE_URL": app.state.config.AUTOMATIC1111_BASE_URL,
-        "AUTOMATIC1111_API_AUTH": app.state.config.AUTOMATIC1111_API_AUTH,
-        "COMFYUI_BASE_URL": app.state.config.COMFYUI_BASE_URL,
-    }
-
-
-@app.post("/url/update")
-async def update_engine_url(
-    form_data: EngineUrlUpdateForm, user=Depends(get_admin_user)
-):
-    if form_data.AUTOMATIC1111_BASE_URL == None:
-        app.state.config.AUTOMATIC1111_BASE_URL = AUTOMATIC1111_BASE_URL
-    else:
-        url = form_data.AUTOMATIC1111_BASE_URL.strip("/")
+@app.get("/config/url/verify")
+async def verify_url(user=Depends(get_admin_user)):
+    if app.state.config.ENGINE == "automatic1111":
         try:
-            r = requests.head(url)
-            app.state.config.AUTOMATIC1111_BASE_URL = url
+            r = requests.get(
+                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
+                headers={"authorization": get_automatic1111_api_auth()},
+            )
+            r.raise_for_status()
+            return True
+        except Exception:
+            app.state.config.ENABLED = False
+            raise HTTPException(status_code=400, detail=ERROR_MESSAGES.INVALID_URL)
+    elif app.state.config.ENGINE == "comfyui":
+        try:
+            r = requests.get(url=f"{app.state.config.COMFYUI_BASE_URL}/object_info")
+            r.raise_for_status()
+            return True
+        except Exception:
+            app.state.config.ENABLED = False
+            raise HTTPException(status_code=400, detail=ERROR_MESSAGES.INVALID_URL)
+    else:
+        return True
+
+
+def set_image_model(model: str):
+    log.info(f"Setting image model to {model}")
+    app.state.config.MODEL = model
+    if app.state.config.ENGINE in ["", "automatic1111"]:
+        api_auth = get_automatic1111_api_auth()
+        r = requests.get(
+            url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
+            headers={"authorization": api_auth},
+        )
+        options = r.json()
+        if model != options["sd_model_checkpoint"]:
+            options["sd_model_checkpoint"] = model
+            r = requests.post(
+                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
+                json=options,
+                headers={"authorization": api_auth},
+            )
+    return app.state.config.MODEL
+
+
+def get_image_model():
+    if app.state.config.ENGINE == "openai":
+        return app.state.config.MODEL if app.state.config.MODEL else "dall-e-2"
+    elif app.state.config.ENGINE == "comfyui":
+        return app.state.config.MODEL if app.state.config.MODEL else ""
+    elif app.state.config.ENGINE == "automatic1111" or app.state.config.ENGINE == "":
+        try:
+            r = requests.get(
+                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
+                headers={"authorization": get_automatic1111_api_auth()},
+            )
+            options = r.json()
+            return options["sd_model_checkpoint"]
         except Exception as e:
+            app.state.config.ENABLED = False
             raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
 
-    if form_data.COMFYUI_BASE_URL == None:
-        app.state.config.COMFYUI_BASE_URL = COMFYUI_BASE_URL
-    else:
-        url = form_data.COMFYUI_BASE_URL.strip("/")
 
-        try:
-            r = requests.head(url)
-            app.state.config.COMFYUI_BASE_URL = url
-        except Exception as e:
-            raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
+class ImageConfigForm(BaseModel):
+    MODEL: str
+    IMAGE_SIZE: str
+    IMAGE_STEPS: int
 
-    if form_data.AUTOMATIC1111_API_AUTH == None:
-        app.state.config.AUTOMATIC1111_API_AUTH = AUTOMATIC1111_API_AUTH
-    else:
-        app.state.config.AUTOMATIC1111_API_AUTH = form_data.AUTOMATIC1111_API_AUTH
 
+@app.get("/image/config")
+async def get_image_config(user=Depends(get_admin_user)):
     return {
-        "AUTOMATIC1111_BASE_URL": app.state.config.AUTOMATIC1111_BASE_URL,
-        "AUTOMATIC1111_API_AUTH": app.state.config.AUTOMATIC1111_API_AUTH,
-        "COMFYUI_BASE_URL": app.state.config.COMFYUI_BASE_URL,
-        "status": True,
+        "MODEL": app.state.config.MODEL,
+        "IMAGE_SIZE": app.state.config.IMAGE_SIZE,
+        "IMAGE_STEPS": app.state.config.IMAGE_STEPS,
     }
 
 
-class OpenAIConfigUpdateForm(BaseModel):
-    url: str
-    key: str
+@app.post("/image/config/update")
+async def update_image_config(form_data: ImageConfigForm, user=Depends(get_admin_user)):
 
+    set_image_model(form_data.MODEL)
 
-@app.get("/openai/config")
-async def get_openai_config(user=Depends(get_admin_user)):
-    return {
-        "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL,
-        "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY,
-    }
-
-
-@app.post("/openai/config/update")
-async def update_openai_config(
-    form_data: OpenAIConfigUpdateForm, user=Depends(get_admin_user)
-):
-    if form_data.key == "":
-        raise HTTPException(status_code=400, detail=ERROR_MESSAGES.API_KEY_NOT_FOUND)
-
-    app.state.config.OPENAI_API_BASE_URL = form_data.url
-    app.state.config.OPENAI_API_KEY = form_data.key
-
-    return {
-        "status": True,
-        "OPENAI_API_BASE_URL": app.state.config.OPENAI_API_BASE_URL,
-        "OPENAI_API_KEY": app.state.config.OPENAI_API_KEY,
-    }
-
-
-class ImageSizeUpdateForm(BaseModel):
-    size: str
-
-
-@app.get("/size")
-async def get_image_size(user=Depends(get_admin_user)):
-    return {"IMAGE_SIZE": app.state.config.IMAGE_SIZE}
-
-
-@app.post("/size/update")
-async def update_image_size(
-    form_data: ImageSizeUpdateForm, user=Depends(get_admin_user)
-):
-    pattern = r"^\d+x\d+$"  # Regular expression pattern
-    if re.match(pattern, form_data.size):
-        app.state.config.IMAGE_SIZE = form_data.size
-        return {
-            "IMAGE_SIZE": app.state.config.IMAGE_SIZE,
-            "status": True,
-        }
+    pattern = r"^\d+x\d+$"
+    if re.match(pattern, form_data.IMAGE_SIZE):
+        app.state.config.IMAGE_SIZE = form_data.IMAGE_SIZE
     else:
         raise HTTPException(
             status_code=400,
             detail=ERROR_MESSAGES.INCORRECT_FORMAT("  (e.g., 512x512)."),
         )
 
-
-class ImageStepsUpdateForm(BaseModel):
-    steps: int
-
-
-@app.get("/steps")
-async def get_image_size(user=Depends(get_admin_user)):
-    return {"IMAGE_STEPS": app.state.config.IMAGE_STEPS}
-
-
-@app.post("/steps/update")
-async def update_image_size(
-    form_data: ImageStepsUpdateForm, user=Depends(get_admin_user)
-):
-    if form_data.steps >= 0:
-        app.state.config.IMAGE_STEPS = form_data.steps
-        return {
-            "IMAGE_STEPS": app.state.config.IMAGE_STEPS,
-            "status": True,
-        }
+    if form_data.IMAGE_STEPS >= 0:
+        app.state.config.IMAGE_STEPS = form_data.IMAGE_STEPS
     else:
         raise HTTPException(
             status_code=400,
             detail=ERROR_MESSAGES.INCORRECT_FORMAT("  (e.g., 50)."),
         )
 
+    return {
+        "MODEL": app.state.config.MODEL,
+        "IMAGE_SIZE": app.state.config.IMAGE_SIZE,
+        "IMAGE_STEPS": app.state.config.IMAGE_STEPS,
+    }
+
 
 @app.get("/models")
 def get_models(user=Depends(get_verified_user)):
@@ -266,18 +313,51 @@ def get_models(user=Depends(get_verified_user)):
                 {"id": "dall-e-3", "name": "DALL·E 3"},
             ]
         elif app.state.config.ENGINE == "comfyui":
-
+            # TODO - get models from comfyui
             r = requests.get(url=f"{app.state.config.COMFYUI_BASE_URL}/object_info")
             info = r.json()
 
-            return list(
-                map(
-                    lambda model: {"id": model, "name": model},
-                    info["CheckpointLoaderSimple"]["input"]["required"]["ckpt_name"][0],
-                )
-            )
+            workflow = json.loads(app.state.config.COMFYUI_WORKFLOW)
+            model_node_id = None
 
-        else:
+            for node in app.state.config.COMFYUI_WORKFLOW_NODES:
+                if node["type"] == "model":
+                    if node["node_ids"]:
+                        model_node_id = node["node_ids"][0]
+                    break
+
+            if model_node_id:
+                model_list_key = None
+
+                print(workflow[model_node_id]["class_type"])
+                for key in info[workflow[model_node_id]["class_type"]]["input"][
+                    "required"
+                ]:
+                    if "_name" in key:
+                        model_list_key = key
+                        break
+
+                if model_list_key:
+                    return list(
+                        map(
+                            lambda model: {"id": model, "name": model},
+                            info[workflow[model_node_id]["class_type"]]["input"][
+                                "required"
+                            ][model_list_key][0],
+                        )
+                    )
+            else:
+                return list(
+                    map(
+                        lambda model: {"id": model, "name": model},
+                        info["CheckpointLoaderSimple"]["input"]["required"][
+                            "ckpt_name"
+                        ][0],
+                    )
+                )
+        elif (
+            app.state.config.ENGINE == "automatic1111" or app.state.config.ENGINE == ""
+        ):
             r = requests.get(
                 url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/sd-models",
                 headers={"authorization": get_automatic1111_api_auth()},
@@ -294,69 +374,11 @@ def get_models(user=Depends(get_verified_user)):
         raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
 
 
-@app.get("/models/default")
-async def get_default_model(user=Depends(get_admin_user)):
-    try:
-        if app.state.config.ENGINE == "openai":
-            return {
-                "model": (
-                    app.state.config.MODEL if app.state.config.MODEL else "dall-e-2"
-                )
-            }
-        elif app.state.config.ENGINE == "comfyui":
-            return {"model": (app.state.config.MODEL if app.state.config.MODEL else "")}
-        else:
-            r = requests.get(
-                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
-                headers={"authorization": get_automatic1111_api_auth()},
-            )
-            options = r.json()
-            return {"model": options["sd_model_checkpoint"]}
-    except Exception as e:
-        app.state.config.ENABLED = False
-        raise HTTPException(status_code=400, detail=ERROR_MESSAGES.DEFAULT(e))
-
-
-class UpdateModelForm(BaseModel):
-    model: str
-
-
-def set_model_handler(model: str):
-    if app.state.config.ENGINE in ["openai", "comfyui"]:
-        app.state.config.MODEL = model
-        return app.state.config.MODEL
-    else:
-        api_auth = get_automatic1111_api_auth()
-        r = requests.get(
-            url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
-            headers={"authorization": api_auth},
-        )
-        options = r.json()
-
-        if model != options["sd_model_checkpoint"]:
-            options["sd_model_checkpoint"] = model
-            r = requests.post(
-                url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/options",
-                json=options,
-                headers={"authorization": api_auth},
-            )
-
-        return options
-
-
-@app.post("/models/default/update")
-def update_default_model(
-    form_data: UpdateModelForm,
-    user=Depends(get_verified_user),
-):
-    return set_model_handler(form_data.model)
-
-
 class GenerateImageForm(BaseModel):
     model: Optional[str] = None
     prompt: str
-    n: int = 1
     size: Optional[str] = None
+    n: int = 1
     negative_prompt: Optional[str] = None
 
 
@@ -398,7 +420,6 @@ def save_url_image(url):
         r = requests.get(url)
         r.raise_for_status()
         if r.headers["content-type"].split("/")[0] == "image":
-
             mime_type = r.headers["content-type"]
             image_format = mimetypes.guess_extension(mime_type)
 
@@ -413,7 +434,7 @@ def save_url_image(url):
                     image_file.write(chunk)
             return image_filename
         else:
-            log.error(f"Url does not point to an image.")
+            log.error("Url does not point to an image.")
             return None
 
     except Exception as e:
@@ -422,7 +443,7 @@ def save_url_image(url):
 
 
 @app.post("/generations")
-def generate_image(
+async def image_generations(
     form_data: GenerateImageForm,
     user=Depends(get_verified_user),
 ):
@@ -431,7 +452,6 @@ def generate_image(
     r = None
     try:
         if app.state.config.ENGINE == "openai":
-
             headers = {}
             headers["Authorization"] = f"Bearer {app.state.config.OPENAI_API_KEY}"
             headers["Content-Type"] = "application/json"
@@ -450,7 +470,9 @@ def generate_image(
                 "response_format": "b64_json",
             }
 
-            r = requests.post(
+            # Use asyncio.to_thread for the requests.post call
+            r = await asyncio.to_thread(
+                requests.post,
                 url=f"{app.state.config.OPENAI_API_BASE_URL}/images/generations",
                 json=data,
                 headers=headers,
@@ -472,7 +494,6 @@ def generate_image(
             return images
 
         elif app.state.config.ENGINE == "comfyui":
-
             data = {
                 "prompt": form_data.prompt,
                 "width": width,
@@ -486,23 +507,20 @@ def generate_image(
             if form_data.negative_prompt is not None:
                 data["negative_prompt"] = form_data.negative_prompt
 
-            if app.state.config.COMFYUI_CFG_SCALE:
-                data["cfg_scale"] = app.state.config.COMFYUI_CFG_SCALE
-
-            if app.state.config.COMFYUI_SAMPLER is not None:
-                data["sampler"] = app.state.config.COMFYUI_SAMPLER
-
-            if app.state.config.COMFYUI_SCHEDULER is not None:
-                data["scheduler"] = app.state.config.COMFYUI_SCHEDULER
-
-            if app.state.config.COMFYUI_SD3 is not None:
-                data["sd3"] = app.state.config.COMFYUI_SD3
-
-            data = ImageGenerationPayload(**data)
-
-            res = comfyui_generate_image(
+            form_data = ComfyUIGenerateImageForm(
+                **{
+                    "workflow": ComfyUIWorkflow(
+                        **{
+                            "workflow": app.state.config.COMFYUI_WORKFLOW,
+                            "nodes": app.state.config.COMFYUI_WORKFLOW_NODES,
+                        }
+                    ),
+                    **data,
+                }
+            )
+            res = await comfyui_generate_image(
                 app.state.config.MODEL,
-                data,
+                form_data,
                 user.id,
                 app.state.config.COMFYUI_BASE_URL,
             )
@@ -516,13 +534,15 @@ def generate_image(
                 file_body_path = IMAGE_CACHE_DIR.joinpath(f"{image_filename}.json")
 
                 with open(file_body_path, "w") as f:
-                    json.dump(data.model_dump(exclude_none=True), f)
+                    json.dump(form_data.model_dump(exclude_none=True), f)
 
             log.debug(f"images: {images}")
             return images
-        else:
+        elif (
+            app.state.config.ENGINE == "automatic1111" or app.state.config.ENGINE == ""
+        ):
             if form_data.model:
-                set_model_handler(form_data.model)
+                set_image_model(form_data.model)
 
             data = {
                 "prompt": form_data.prompt,
@@ -537,14 +557,24 @@ def generate_image(
             if form_data.negative_prompt is not None:
                 data["negative_prompt"] = form_data.negative_prompt
 
-            r = requests.post(
+            if app.state.config.AUTOMATIC1111_CFG_SCALE:
+                data["cfg_scale"] = app.state.config.AUTOMATIC1111_CFG_SCALE
+
+            if app.state.config.AUTOMATIC1111_SAMPLER:
+                data["sampler_name"] = app.state.config.AUTOMATIC1111_SAMPLER
+
+            if app.state.config.AUTOMATIC1111_SCHEDULER:
+                data["scheduler"] = app.state.config.AUTOMATIC1111_SCHEDULER
+
+            # Use asyncio.to_thread for the requests.post call
+            r = await asyncio.to_thread(
+                requests.post,
                 url=f"{app.state.config.AUTOMATIC1111_BASE_URL}/sdapi/v1/txt2img",
                 json=data,
                 headers={"authorization": get_automatic1111_api_auth()},
             )
 
             res = r.json()
-
             log.debug(f"res: {res}")
 
             images = []
@@ -558,10 +588,8 @@ def generate_image(
                     json.dump({**data, "info": res["info"]}, f)
 
             return images
-
     except Exception as e:
         error = e
-
         if r != None:
             data = r.json()
             if "error" in data:

backend/open_webui/apps/images/utils/comfyui.py

@@ -0,0 +1,174 @@
+import asyncio
+import json
+import logging
+import random
+import urllib.parse
+import urllib.request
+from typing import Optional
+
+import websocket  # NOTE: websocket-client (https://github.com/websocket-client/websocket-client)
+from open_webui.env import SRC_LOG_LEVELS
+from pydantic import BaseModel
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["COMFYUI"])
+
+default_headers = {"User-Agent": "Mozilla/5.0"}
+
+
+def queue_prompt(prompt, client_id, base_url):
+    log.info("queue_prompt")
+    p = {"prompt": prompt, "client_id": client_id}
+    data = json.dumps(p).encode("utf-8")
+    log.debug(f"queue_prompt data: {data}")
+    try:
+        req = urllib.request.Request(
+            f"{base_url}/prompt", data=data, headers=default_headers
+        )
+        response = urllib.request.urlopen(req).read()
+        return json.loads(response)
+    except Exception as e:
+        log.exception(f"Error while queuing prompt: {e}")
+        raise e
+
+
+def get_image(filename, subfolder, folder_type, base_url):
+    log.info("get_image")
+    data = {"filename": filename, "subfolder": subfolder, "type": folder_type}
+    url_values = urllib.parse.urlencode(data)
+    req = urllib.request.Request(
+        f"{base_url}/view?{url_values}", headers=default_headers
+    )
+    with urllib.request.urlopen(req) as response:
+        return response.read()
+
+
+def get_image_url(filename, subfolder, folder_type, base_url):
+    log.info("get_image")
+    data = {"filename": filename, "subfolder": subfolder, "type": folder_type}
+    url_values = urllib.parse.urlencode(data)
+    return f"{base_url}/view?{url_values}"
+
+
+def get_history(prompt_id, base_url):
+    log.info("get_history")
+
+    req = urllib.request.Request(
+        f"{base_url}/history/{prompt_id}", headers=default_headers
+    )
+    with urllib.request.urlopen(req) as response:
+        return json.loads(response.read())
+
+
+def get_images(ws, prompt, client_id, base_url):
+    prompt_id = queue_prompt(prompt, client_id, base_url)["prompt_id"]
+    output_images = []
+    while True:
+        out = ws.recv()
+        if isinstance(out, str):
+            message = json.loads(out)
+            if message["type"] == "executing":
+                data = message["data"]
+                if data["node"] is None and data["prompt_id"] == prompt_id:
+                    break  # Execution is done
+        else:
+            continue  # previews are binary data
+
+    history = get_history(prompt_id, base_url)[prompt_id]
+    for o in history["outputs"]:
+        for node_id in history["outputs"]:
+            node_output = history["outputs"][node_id]
+            if "images" in node_output:
+                for image in node_output["images"]:
+                    url = get_image_url(
+                        image["filename"], image["subfolder"], image["type"], base_url
+                    )
+                    output_images.append({"url": url})
+    return {"data": output_images}
+
+
+class ComfyUINodeInput(BaseModel):
+    type: Optional[str] = None
+    node_ids: list[str] = []
+    key: Optional[str] = "text"
+    value: Optional[str] = None
+
+
+class ComfyUIWorkflow(BaseModel):
+    workflow: str
+    nodes: list[ComfyUINodeInput]
+
+
+class ComfyUIGenerateImageForm(BaseModel):
+    workflow: ComfyUIWorkflow
+
+    prompt: str
+    negative_prompt: Optional[str] = None
+    width: int
+    height: int
+    n: int = 1
+
+    steps: Optional[int] = None
+    seed: Optional[int] = None
+
+
+async def comfyui_generate_image(
+    model: str, payload: ComfyUIGenerateImageForm, client_id, base_url
+):
+    ws_url = base_url.replace("http://", "ws://").replace("https://", "wss://")
+    workflow = json.loads(payload.workflow.workflow)
+
+    for node in payload.workflow.nodes:
+        if node.type:
+            if node.type == "model":
+                for node_id in node.node_ids:
+                    workflow[node_id]["inputs"][node.key] = model
+            elif node.type == "prompt":
+                for node_id in node.node_ids:
+                    workflow[node_id]["inputs"]["text"] = payload.prompt
+            elif node.type == "negative_prompt":
+                for node_id in node.node_ids:
+                    workflow[node_id]["inputs"]["text"] = payload.negative_prompt
+            elif node.type == "width":
+                for node_id in node.node_ids:
+                    workflow[node_id]["inputs"]["width"] = payload.width
+            elif node.type == "height":
+                for node_id in node.node_ids:
+                    workflow[node_id]["inputs"]["height"] = payload.height
+            elif node.type == "n":
+                for node_id in node.node_ids:
+                    workflow[node_id]["inputs"]["batch_size"] = payload.n
+            elif node.type == "steps":
+                for node_id in node.node_ids:
+                    workflow[node_id]["inputs"]["steps"] = payload.steps
+            elif node.type == "seed":
+                seed = (
+                    payload.seed
+                    if payload.seed
+                    else random.randint(0, 18446744073709551614)
+                )
+                for node_id in node.node_ids:
+                    workflow[node_id]["inputs"][node.key] = seed
+        else:
+            for node_id in node.node_ids:
+                workflow[node_id]["inputs"][node.key] = node.value
+
+    try:
+        ws = websocket.WebSocket()
+        ws.connect(f"{ws_url}/ws?clientId={client_id}")
+        log.info("WebSocket connection established.")
+    except Exception as e:
+        log.exception(f"Failed to connect to WebSocket server: {e}")
+        return None
+
+    try:
+        log.info("Sending workflow to WebSocket server.")
+        log.info(f"Workflow: {workflow}")
+        images = await asyncio.to_thread(get_images, ws, workflow, client_id, base_url)
+    except Exception as e:
+        log.exception(f"Error while receiving images: {e}")
+        images = None
+
+    ws.close()
+
+    return images

backend/open_webui/apps/ollama/main.py

@@ -1,59 +1,44 @@
-from fastapi import (
-    FastAPI,
-    Request,
-    Response,
-    HTTPException,
-    Depends,
-    status,
-    UploadFile,
-    File,
-    BackgroundTasks,
-)
-from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import StreamingResponse
-from fastapi.concurrency import run_in_threadpool
-
-from pydantic import BaseModel, ConfigDict
-
-import os
-import re
-import copy
-import random
-import requests
-import json
-import uuid
-import aiohttp
 import asyncio
+import json
 import logging
+import os
+import random
+import re
 import time
+from typing import Optional, Union
 from urllib.parse import urlparse
-from typing import Optional, List, Union
 
-from starlette.background import BackgroundTask
-
-from apps.webui.models.models import Models
-from apps.webui.models.users import Users
-from constants import ERROR_MESSAGES
-from utils.utils import (
-    decode_token,
-    get_current_user,
-    get_verified_user,
-    get_admin_user,
-)
-from utils.task import prompt_template
-
-
-from config import (
-    SRC_LOG_LEVELS,
-    OLLAMA_BASE_URLS,
-    ENABLE_OLLAMA_API,
+import aiohttp
+import requests
+from open_webui.apps.webui.models.models import Models
+from open_webui.config import (
     AIOHTTP_CLIENT_TIMEOUT,
+    CORS_ALLOW_ORIGIN,
     ENABLE_MODEL_FILTER,
+    ENABLE_OLLAMA_API,
     MODEL_FILTER_LIST,
+    OLLAMA_BASE_URLS,
     UPLOAD_DIR,
     AppConfig,
 )
-from utils.misc import calculate_sha256, add_or_update_system_message
+from open_webui.constants import ERROR_MESSAGES
+from open_webui.env import SRC_LOG_LEVELS
+from fastapi import Depends, FastAPI, File, HTTPException, Request, UploadFile
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import StreamingResponse
+from pydantic import BaseModel, ConfigDict
+from starlette.background import BackgroundTask
+
+
+from open_webui.utils.misc import (
+    calculate_sha256,
+)
+from open_webui.utils.payload import (
+    apply_model_params_to_body_ollama,
+    apply_model_params_to_body_openai,
+    apply_model_system_prompt_to_body,
+)
+from open_webui.utils.utils import get_admin_user, get_verified_user
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["OLLAMA"])
@@ -61,7 +46,7 @@ log.setLevel(SRC_LOG_LEVELS["OLLAMA"])
 app = FastAPI()
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=["*"],
+    allow_origins=CORS_ALLOW_ORIGIN,
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],
@@ -120,7 +105,7 @@ async def get_ollama_api_urls(user=Depends(get_admin_user)):
 
 
 class UrlUpdateForm(BaseModel):
-    urls: List[str]
+    urls: list[str]
 
 
 @app.post("/urls/update")
@@ -153,21 +138,38 @@ async def cleanup_response(
         await session.close()
 
 
-async def post_streaming_url(url: str, payload: str):
+async def post_streaming_url(
+    url: str, payload: Union[str, bytes], stream: bool = True, content_type=None
+):
     r = None
     try:
         session = aiohttp.ClientSession(
             trust_env=True, timeout=aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT)
         )
-        r = await session.post(url, data=payload)
+        r = await session.post(
+            url,
+            data=payload,
+            headers={"Content-Type": "application/json"},
+        )
         r.raise_for_status()
 
-        return StreamingResponse(
-            r.content,
-            status_code=r.status,
-            headers=dict(r.headers),
-            background=BackgroundTask(cleanup_response, response=r, session=session),
-        )
+        if stream:
+            headers = dict(r.headers)
+            if content_type:
+                headers["Content-Type"] = content_type
+            return StreamingResponse(
+                r.content,
+                status_code=r.status,
+                headers=headers,
+                background=BackgroundTask(
+                    cleanup_response, response=r, session=session
+                ),
+            )
+        else:
+            res = await r.json()
+            await cleanup_response(r, session)
+            return res
+
     except Exception as e:
         error_detail = "Open WebUI: Server Connection Error"
         if r is not None:
@@ -175,7 +177,7 @@ async def post_streaming_url(url: str, payload: str):
                 res = await r.json()
                 if "error" in res:
                     error_detail = f"Ollama: {res['error']}"
-            except:
+            except Exception:
                 error_detail = f"Ollama: {e}"
 
         raise HTTPException(
@@ -230,7 +232,7 @@ async def get_all_models():
 async def get_ollama_tags(
     url_idx: Optional[int] = None, user=Depends(get_verified_user)
 ):
-    if url_idx == None:
+    if url_idx is None:
         models = await get_all_models()
 
         if app.state.config.ENABLE_MODEL_FILTER:
@@ -261,7 +263,7 @@ async def get_ollama_tags(
                     res = r.json()
                     if "error" in res:
                         error_detail = f"Ollama: {res['error']}"
-                except:
+                except Exception:
                     error_detail = f"Ollama: {e}"
 
             raise HTTPException(
@@ -274,8 +276,7 @@ async def get_ollama_tags(
 @app.get("/api/version/{url_idx}")
 async def get_ollama_versions(url_idx: Optional[int] = None):
     if app.state.config.ENABLE_OLLAMA_API:
-        if url_idx == None:
-
+        if url_idx is None:
             # returns lowest version
             tasks = [
                 fetch_url(f"{url}/api/version")
@@ -315,7 +316,7 @@ async def get_ollama_versions(url_idx: Optional[int] = None):
                         res = r.json()
                         if "error" in res:
                             error_detail = f"Ollama: {res['error']}"
-                    except:
+                    except Exception:
                         error_detail = f"Ollama: {e}"
 
                 raise HTTPException(
@@ -338,8 +339,6 @@ async def pull_model(
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
-    r = None
-
     # Admin should be able to pull models from any source
     payload = {**form_data.model_dump(exclude_none=True), "insecure": True}
 
@@ -359,7 +358,7 @@ async def push_model(
     url_idx: Optional[int] = None,
     user=Depends(get_admin_user),
 ):
-    if url_idx == None:
+    if url_idx is None:
         if form_data.name in app.state.MODELS:
             url_idx = app.state.MODELS[form_data.name]["urls"][0]
         else:
@@ -409,7 +408,7 @@ async def copy_model(
     url_idx: Optional[int] = None,
     user=Depends(get_admin_user),
 ):
-    if url_idx == None:
+    if url_idx is None:
         if form_data.source in app.state.MODELS:
             url_idx = app.state.MODELS[form_data.source]["urls"][0]
         else:
@@ -420,13 +419,14 @@ async def copy_model(
 
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
+    r = requests.request(
+        method="POST",
+        url=f"{url}/api/copy",
+        headers={"Content-Type": "application/json"},
+        data=form_data.model_dump_json(exclude_none=True).encode(),
+    )
 
     try:
-        r = requests.request(
-            method="POST",
-            url=f"{url}/api/copy",
-            data=form_data.model_dump_json(exclude_none=True).encode(),
-        )
         r.raise_for_status()
 
         log.debug(f"r.text: {r.text}")
@@ -440,7 +440,7 @@ async def copy_model(
                 res = r.json()
                 if "error" in res:
                     error_detail = f"Ollama: {res['error']}"
-            except:
+            except Exception:
                 error_detail = f"Ollama: {e}"
 
         raise HTTPException(
@@ -456,7 +456,7 @@ async def delete_model(
     url_idx: Optional[int] = None,
     user=Depends(get_admin_user),
 ):
-    if url_idx == None:
+    if url_idx is None:
         if form_data.name in app.state.MODELS:
             url_idx = app.state.MODELS[form_data.name]["urls"][0]
         else:
@@ -468,12 +468,13 @@ async def delete_model(
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
+    r = requests.request(
+        method="DELETE",
+        url=f"{url}/api/delete",
+        headers={"Content-Type": "application/json"},
+        data=form_data.model_dump_json(exclude_none=True).encode(),
+    )
     try:
-        r = requests.request(
-            method="DELETE",
-            url=f"{url}/api/delete",
-            data=form_data.model_dump_json(exclude_none=True).encode(),
-        )
         r.raise_for_status()
 
         log.debug(f"r.text: {r.text}")
@@ -487,7 +488,7 @@ async def delete_model(
                 res = r.json()
                 if "error" in res:
                     error_detail = f"Ollama: {res['error']}"
-            except:
+            except Exception:
                 error_detail = f"Ollama: {e}"
 
         raise HTTPException(
@@ -508,12 +509,13 @@ async def show_model_info(form_data: ModelNameForm, user=Depends(get_verified_us
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
+    r = requests.request(
+        method="POST",
+        url=f"{url}/api/show",
+        headers={"Content-Type": "application/json"},
+        data=form_data.model_dump_json(exclude_none=True).encode(),
+    )
     try:
-        r = requests.request(
-            method="POST",
-            url=f"{url}/api/show",
-            data=form_data.model_dump_json(exclude_none=True).encode(),
-        )
         r.raise_for_status()
 
         return r.json()
@@ -525,7 +527,7 @@ async def show_model_info(form_data: ModelNameForm, user=Depends(get_verified_us
                 res = r.json()
                 if "error" in res:
                     error_detail = f"Ollama: {res['error']}"
-            except:
+            except Exception:
                 error_detail = f"Ollama: {e}"
 
         raise HTTPException(
@@ -541,14 +543,14 @@ class GenerateEmbeddingsForm(BaseModel):
     keep_alive: Optional[Union[int, str]] = None
 
 
-@app.post("/api/embeddings")
-@app.post("/api/embeddings/{url_idx}")
+@app.post("/api/embed")
+@app.post("/api/embed/{url_idx}")
 async def generate_embeddings(
     form_data: GenerateEmbeddingsForm,
     url_idx: Optional[int] = None,
     user=Depends(get_verified_user),
 ):
-    if url_idx == None:
+    if url_idx is None:
         model = form_data.model
 
         if ":" not in model:
@@ -565,12 +567,13 @@ async def generate_embeddings(
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
+    r = requests.request(
+        method="POST",
+        url=f"{url}/api/embed",
+        headers={"Content-Type": "application/json"},
+        data=form_data.model_dump_json(exclude_none=True).encode(),
+    )
     try:
-        r = requests.request(
-            method="POST",
-            url=f"{url}/api/embeddings",
-            data=form_data.model_dump_json(exclude_none=True).encode(),
-        )
         r.raise_for_status()
 
         return r.json()
@@ -582,7 +585,58 @@ async def generate_embeddings(
                 res = r.json()
                 if "error" in res:
                     error_detail = f"Ollama: {res['error']}"
-            except:
+            except Exception:
+                error_detail = f"Ollama: {e}"
+
+        raise HTTPException(
+            status_code=r.status_code if r else 500,
+            detail=error_detail,
+        )
+
+
+@app.post("/api/embeddings")
+@app.post("/api/embeddings/{url_idx}")
+async def generate_embeddings(
+    form_data: GenerateEmbeddingsForm,
+    url_idx: Optional[int] = None,
+    user=Depends(get_verified_user),
+):
+    if url_idx is None:
+        model = form_data.model
+
+        if ":" not in model:
+            model = f"{model}:latest"
+
+        if model in app.state.MODELS:
+            url_idx = random.choice(app.state.MODELS[model]["urls"])
+        else:
+            raise HTTPException(
+                status_code=400,
+                detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
+            )
+
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
+    log.info(f"url: {url}")
+
+    r = requests.request(
+        method="POST",
+        url=f"{url}/api/embeddings",
+        headers={"Content-Type": "application/json"},
+        data=form_data.model_dump_json(exclude_none=True).encode(),
+    )
+    try:
+        r.raise_for_status()
+
+        return r.json()
+    except Exception as e:
+        log.exception(e)
+        error_detail = "Open WebUI: Server Connection Error"
+        if r is not None:
+            try:
+                res = r.json()
+                if "error" in res:
+                    error_detail = f"Ollama: {res['error']}"
+            except Exception:
                 error_detail = f"Ollama: {e}"
 
         raise HTTPException(
@@ -595,10 +649,9 @@ def generate_ollama_embeddings(
     form_data: GenerateEmbeddingsForm,
     url_idx: Optional[int] = None,
 ):
-
     log.info(f"generate_ollama_embeddings {form_data}")
 
-    if url_idx == None:
+    if url_idx is None:
         model = form_data.model
 
         if ":" not in model:
@@ -615,12 +668,13 @@ def generate_ollama_embeddings(
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
     log.info(f"url: {url}")
 
+    r = requests.request(
+        method="POST",
+        url=f"{url}/api/embeddings",
+        headers={"Content-Type": "application/json"},
+        data=form_data.model_dump_json(exclude_none=True).encode(),
+    )
     try:
-        r = requests.request(
-            method="POST",
-            url=f"{url}/api/embeddings",
-            data=form_data.model_dump_json(exclude_none=True).encode(),
-        )
         r.raise_for_status()
 
         data = r.json()
@@ -630,7 +684,7 @@ def generate_ollama_embeddings(
         if "embedding" in data:
             return data["embedding"]
         else:
-            raise "Something went wrong :/"
+            raise Exception("Something went wrong :/")
     except Exception as e:
         log.exception(e)
         error_detail = "Open WebUI: Server Connection Error"
@@ -639,16 +693,16 @@ def generate_ollama_embeddings(
                 res = r.json()
                 if "error" in res:
                     error_detail = f"Ollama: {res['error']}"
-            except:
+            except Exception:
                 error_detail = f"Ollama: {e}"
 
-        raise error_detail
+        raise Exception(error_detail)
 
 
 class GenerateCompletionForm(BaseModel):
     model: str
     prompt: str
-    images: Optional[List[str]] = None
+    images: Optional[list[str]] = None
     format: Optional[str] = None
     options: Optional[dict] = None
     system: Optional[str] = None
@@ -666,8 +720,7 @@ async def generate_completion(
     url_idx: Optional[int] = None,
     user=Depends(get_verified_user),
 ):
-
-    if url_idx == None:
+    if url_idx is None:
         model = form_data.model
 
         if ":" not in model:
@@ -692,12 +745,12 @@ async def generate_completion(
 class ChatMessage(BaseModel):
     role: str
     content: str
-    images: Optional[List[str]] = None
+    images: Optional[list[str]] = None
 
 
 class GenerateChatCompletionForm(BaseModel):
     model: str
-    messages: List[ChatMessage]
+    messages: list[ChatMessage]
     format: Optional[str] = None
     options: Optional[dict] = None
     template: Optional[str] = None
@@ -705,6 +758,18 @@ class GenerateChatCompletionForm(BaseModel):
     keep_alive: Optional[Union[int, str]] = None
 
 
+def get_ollama_url(url_idx: Optional[int], model: str):
+    if url_idx is None:
+        if model not in app.state.MODELS:
+            raise HTTPException(
+                status_code=400,
+                detail=ERROR_MESSAGES.MODEL_NOT_FOUND(model),
+            )
+        url_idx = random.choice(app.state.MODELS[model]["urls"])
+    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
+    return url
+
+
 @app.post("/api/chat")
 @app.post("/api/chat/{url_idx}")
 async def generate_chat_completion(
@@ -712,149 +777,50 @@ async def generate_chat_completion(
     url_idx: Optional[int] = None,
     user=Depends(get_verified_user),
 ):
-
-    log.debug(
-        "form_data.model_dump_json(exclude_none=True).encode(): {0} ".format(
-            form_data.model_dump_json(exclude_none=True).encode()
-        )
-    )
-
-    payload = {
-        **form_data.model_dump(exclude_none=True),
-    }
+    payload = {**form_data.model_dump(exclude_none=True)}
+    log.debug(f"{payload = }")
+    if "metadata" in payload:
+        del payload["metadata"]
 
     model_id = form_data.model
+
+    if app.state.config.ENABLE_MODEL_FILTER:
+        if user.role == "user" and model_id not in app.state.config.MODEL_FILTER_LIST:
+            raise HTTPException(
+                status_code=403,
+                detail="Model not found",
+            )
+
     model_info = Models.get_model_by_id(model_id)
 
     if model_info:
         if model_info.base_model_id:
             payload["model"] = model_info.base_model_id
 
-        model_info.params = model_info.params.model_dump()
+        params = model_info.params.model_dump()
 
-        if model_info.params:
-            payload["options"] = {}
+        if params:
+            if payload.get("options") is None:
+                payload["options"] = {}
 
-            if model_info.params.get("mirostat", None):
-                payload["options"]["mirostat"] = model_info.params.get("mirostat", None)
-
-            if model_info.params.get("mirostat_eta", None):
-                payload["options"]["mirostat_eta"] = model_info.params.get(
-                    "mirostat_eta", None
-                )
-
-            if model_info.params.get("mirostat_tau", None):
-
-                payload["options"]["mirostat_tau"] = model_info.params.get(
-                    "mirostat_tau", None
-                )
-
-            if model_info.params.get("num_ctx", None):
-                payload["options"]["num_ctx"] = model_info.params.get("num_ctx", None)
-
-            if model_info.params.get("num_batch", None):
-                payload["options"]["num_batch"] = model_info.params.get(
-                    "num_batch", None
-                )
-
-            if model_info.params.get("num_keep", None):
-                payload["options"]["num_keep"] = model_info.params.get("num_keep", None)
-
-            if model_info.params.get("repeat_last_n", None):
-                payload["options"]["repeat_last_n"] = model_info.params.get(
-                    "repeat_last_n", None
-                )
-
-            if model_info.params.get("frequency_penalty", None):
-                payload["options"]["repeat_penalty"] = model_info.params.get(
-                    "frequency_penalty", None
-                )
-
-            if model_info.params.get("temperature", None) is not None:
-                payload["options"]["temperature"] = model_info.params.get(
-                    "temperature", None
-                )
-
-            if model_info.params.get("seed", None):
-                payload["options"]["seed"] = model_info.params.get("seed", None)
-
-            if model_info.params.get("stop", None):
-                payload["options"]["stop"] = (
-                    [
-                        bytes(stop, "utf-8").decode("unicode_escape")
-                        for stop in model_info.params["stop"]
-                    ]
-                    if model_info.params.get("stop", None)
-                    else None
-                )
-
-            if model_info.params.get("tfs_z", None):
-                payload["options"]["tfs_z"] = model_info.params.get("tfs_z", None)
-
-            if model_info.params.get("max_tokens", None):
-                payload["options"]["num_predict"] = model_info.params.get(
-                    "max_tokens", None
-                )
-
-            if model_info.params.get("top_k", None):
-                payload["options"]["top_k"] = model_info.params.get("top_k", None)
-
-            if model_info.params.get("top_p", None):
-                payload["options"]["top_p"] = model_info.params.get("top_p", None)
-
-            if model_info.params.get("use_mmap", None):
-                payload["options"]["use_mmap"] = model_info.params.get("use_mmap", None)
-
-            if model_info.params.get("use_mlock", None):
-                payload["options"]["use_mlock"] = model_info.params.get(
-                    "use_mlock", None
-                )
-
-            if model_info.params.get("num_thread", None):
-                payload["options"]["num_thread"] = model_info.params.get(
-                    "num_thread", None
-                )
-
-        system = model_info.params.get("system", None)
-        if system:
-            # Check if the payload already has a system message
-            # If not, add a system message to the payload
-            system = prompt_template(
-                system,
-                **(
-                    {
-                        "user_name": user.name,
-                        "user_location": (
-                            user.info.get("location") if user.info else None
-                        ),
-                    }
-                    if user
-                    else {}
-                ),
+            payload["options"] = apply_model_params_to_body_ollama(
+                params, payload["options"]
             )
+            payload = apply_model_system_prompt_to_body(params, payload, user)
 
-            if payload.get("messages"):
-                payload["messages"] = add_or_update_system_message(
-                    system, payload["messages"]
-                )
+    if ":" not in payload["model"]:
+        payload["model"] = f"{payload['model']}:latest"
 
-    if url_idx == None:
-        if ":" not in payload["model"]:
-            payload["model"] = f"{payload['model']}:latest"
-
-        if payload["model"] in app.state.MODELS:
-            url_idx = random.choice(app.state.MODELS[payload["model"]]["urls"])
-        else:
-            raise HTTPException(
-                status_code=400,
-                detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
-            )
-
-    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
+    url = get_ollama_url(url_idx, payload["model"])
     log.info(f"url: {url}")
     log.debug(payload)
 
-    return await post_streaming_url(f"{url}/api/chat", json.dumps(payload))
+    return await post_streaming_url(
+        f"{url}/api/chat",
+        json.dumps(payload),
+        stream=form_data.stream,
+        content_type="application/x-ndjson",
+    )
 
 
 # TODO: we should update this part once Ollama supports other types
@@ -872,7 +838,7 @@ class OpenAIChatMessage(BaseModel):
 
 class OpenAIChatCompletionForm(BaseModel):
     model: str
-    messages: List[OpenAIChatMessage]
+    messages: list[OpenAIChatMessage]
 
     model_config = ConfigDict(extra="allow")
 
@@ -884,86 +850,43 @@ async def generate_openai_chat_completion(
     url_idx: Optional[int] = None,
     user=Depends(get_verified_user),
 ):
-    form_data = OpenAIChatCompletionForm(**form_data)
+    completion_form = OpenAIChatCompletionForm(**form_data)
+    payload = {**completion_form.model_dump(exclude_none=True, exclude=["metadata"])}
+    if "metadata" in payload:
+        del payload["metadata"]
 
-    payload = {
-        **form_data.model_dump(exclude_none=True),
-    }
+    model_id = completion_form.model
+
+    if app.state.config.ENABLE_MODEL_FILTER:
+        if user.role == "user" and model_id not in app.state.config.MODEL_FILTER_LIST:
+            raise HTTPException(
+                status_code=403,
+                detail="Model not found",
+            )
 
-    model_id = form_data.model
     model_info = Models.get_model_by_id(model_id)
 
     if model_info:
         if model_info.base_model_id:
             payload["model"] = model_info.base_model_id
 
-        model_info.params = model_info.params.model_dump()
+        params = model_info.params.model_dump()
 
-        if model_info.params:
-            payload["temperature"] = model_info.params.get("temperature", None)
-            payload["top_p"] = model_info.params.get("top_p", None)
-            payload["max_tokens"] = model_info.params.get("max_tokens", None)
-            payload["frequency_penalty"] = model_info.params.get(
-                "frequency_penalty", None
-            )
-            payload["seed"] = model_info.params.get("seed", None)
-            payload["stop"] = (
-                [
-                    bytes(stop, "utf-8").decode("unicode_escape")
-                    for stop in model_info.params["stop"]
-                ]
-                if model_info.params.get("stop", None)
-                else None
-            )
+        if params:
+            payload = apply_model_params_to_body_openai(params, payload)
+            payload = apply_model_system_prompt_to_body(params, payload, user)
 
-        system = model_info.params.get("system", None)
+    if ":" not in payload["model"]:
+        payload["model"] = f"{payload['model']}:latest"
 
-        if system:
-            system = prompt_template(
-                system,
-                **(
-                    {
-                        "user_name": user.name,
-                        "user_location": (
-                            user.info.get("location") if user.info else None
-                        ),
-                    }
-                    if user
-                    else {}
-                ),
-            )
-            # Check if the payload already has a system message
-            # If not, add a system message to the payload
-            if payload.get("messages"):
-                for message in payload["messages"]:
-                    if message.get("role") == "system":
-                        message["content"] = system + message["content"]
-                        break
-                else:
-                    payload["messages"].insert(
-                        0,
-                        {
-                            "role": "system",
-                            "content": system,
-                        },
-                    )
-
-    if url_idx == None:
-        if ":" not in payload["model"]:
-            payload["model"] = f"{payload['model']}:latest"
-
-        if payload["model"] in app.state.MODELS:
-            url_idx = random.choice(app.state.MODELS[payload["model"]]["urls"])
-        else:
-            raise HTTPException(
-                status_code=400,
-                detail=ERROR_MESSAGES.MODEL_NOT_FOUND(form_data.model),
-            )
-
-    url = app.state.config.OLLAMA_BASE_URLS[url_idx]
+    url = get_ollama_url(url_idx, payload["model"])
     log.info(f"url: {url}")
 
-    return await post_streaming_url(f"{url}/v1/chat/completions", json.dumps(payload))
+    return await post_streaming_url(
+        f"{url}/v1/chat/completions",
+        json.dumps(payload),
+        stream=payload.get("stream", False),
+    )
 
 
 @app.get("/v1/models")
@@ -972,7 +895,7 @@ async def get_openai_models(
     url_idx: Optional[int] = None,
     user=Depends(get_verified_user),
 ):
-    if url_idx == None:
+    if url_idx is None:
         models = await get_all_models()
 
         if app.state.config.ENABLE_MODEL_FILTER:
@@ -1027,7 +950,7 @@ async def get_openai_models(
                     res = r.json()
                     if "error" in res:
                         error_detail = f"Ollama: {res['error']}"
-                except:
+                except Exception:
                     error_detail = f"Ollama: {e}"
 
             raise HTTPException(
@@ -1053,7 +976,6 @@ def parse_huggingface_url(hf_url):
         path_components = parsed_url.path.split("/")
 
         # Extract the desired output
-        user_repo = "/".join(path_components[1:3])
         model_file = path_components[-1]
 
         return model_file
@@ -1118,7 +1040,6 @@ async def download_model(
     url_idx: Optional[int] = None,
     user=Depends(get_admin_user),
 ):
-
     allowed_hosts = ["https://huggingface.co/", "https://github.com/"]
 
     if not any(form_data.url.startswith(host) for host in allowed_hosts):
@@ -1127,7 +1048,7 @@ async def download_model(
             detail="Invalid file_url. Only URLs from allowed hosts are permitted.",
         )
 
-    if url_idx == None:
+    if url_idx is None:
         url_idx = 0
     url = app.state.config.OLLAMA_BASE_URLS[url_idx]
 
@@ -1150,7 +1071,7 @@ def upload_model(
     url_idx: Optional[int] = None,
     user=Depends(get_admin_user),
 ):
-    if url_idx == None:
+    if url_idx is None:
         url_idx = 0
     ollama_url = app.state.config.OLLAMA_BASE_URLS[url_idx]
 

backend/open_webui/apps/openai/main.py

@@ -1,42 +1,39 @@
-from fastapi import FastAPI, Request, Response, HTTPException, Depends
-from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import StreamingResponse, JSONResponse, FileResponse
-
-import requests
-import aiohttp
 import asyncio
+import hashlib
 import json
 import logging
+from pathlib import Path
+from typing import Literal, Optional, overload
 
+import aiohttp
+import requests
+from open_webui.apps.webui.models.models import Models
+from open_webui.config import (
+    AIOHTTP_CLIENT_TIMEOUT,
+    CACHE_DIR,
+    CORS_ALLOW_ORIGIN,
+    ENABLE_MODEL_FILTER,
+    ENABLE_OPENAI_API,
+    MODEL_FILTER_LIST,
+    OPENAI_API_BASE_URLS,
+    OPENAI_API_KEYS,
+    AppConfig,
+)
+from open_webui.constants import ERROR_MESSAGES
+from open_webui.env import SRC_LOG_LEVELS
+from fastapi import Depends, FastAPI, HTTPException, Request
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import FileResponse, StreamingResponse
 from pydantic import BaseModel
 from starlette.background import BackgroundTask
 
-from apps.webui.models.models import Models
-from apps.webui.models.users import Users
-from constants import ERROR_MESSAGES
-from utils.utils import (
-    decode_token,
-    get_verified_user,
-    get_verified_user,
-    get_admin_user,
+
+from open_webui.utils.payload import (
+    apply_model_params_to_body_openai,
+    apply_model_system_prompt_to_body,
 )
-from utils.task import prompt_template
 
-from config import (
-    SRC_LOG_LEVELS,
-    ENABLE_OPENAI_API,
-    OPENAI_API_BASE_URLS,
-    OPENAI_API_KEYS,
-    CACHE_DIR,
-    ENABLE_MODEL_FILTER,
-    MODEL_FILTER_LIST,
-    AppConfig,
-)
-from typing import List, Optional
-
-
-import hashlib
-from pathlib import Path
+from open_webui.utils.utils import get_admin_user, get_verified_user
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["OPENAI"])
@@ -44,7 +41,7 @@ log.setLevel(SRC_LOG_LEVELS["OPENAI"])
 app = FastAPI()
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=["*"],
+    allow_origins=CORS_ALLOW_ORIGIN,
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],
@@ -67,8 +64,6 @@ app.state.MODELS = {}
 async def check_url(request: Request, call_next):
     if len(app.state.MODELS) == 0:
         await get_all_models()
-    else:
-        pass
 
     response = await call_next(request)
     return response
@@ -90,11 +85,11 @@ async def update_config(form_data: OpenAIConfigForm, user=Depends(get_admin_user
 
 
 class UrlsUpdateForm(BaseModel):
-    urls: List[str]
+    urls: list[str]
 
 
 class KeysUpdateForm(BaseModel):
-    keys: List[str]
+    keys: list[str]
 
 
 @app.get("/urls")
@@ -173,7 +168,7 @@ async def speech(request: Request, user=Depends(get_verified_user)):
                     res = r.json()
                     if "error" in res:
                         error_detail = f"External: {res['error']}"
-                except:
+                except Exception:
                     error_detail = f"External: {e}"
 
             raise HTTPException(
@@ -225,71 +220,85 @@ def merge_models_lists(model_lists):
                     for model in models
                     if "api.openai.com"
                     not in app.state.config.OPENAI_API_BASE_URLS[idx]
-                    or "gpt" in model["id"]
+                    or not any(
+                        name in model["id"]
+                        for name in [
+                            "babbage",
+                            "dall-e",
+                            "davinci",
+                            "embedding",
+                            "tts",
+                            "whisper",
+                        ]
+                    )
                 ]
             )
 
     return merged_list
 
 
-async def get_all_models(raw: bool = False):
+def is_openai_api_disabled():
+    api_keys = app.state.config.OPENAI_API_KEYS
+    no_keys = len(api_keys) == 1 and api_keys[0] == ""
+    return no_keys or not app.state.config.ENABLE_OPENAI_API
+
+
+async def get_all_models_raw() -> list:
+    if is_openai_api_disabled():
+        return []
+
+    # Check if API KEYS length is same than API URLS length
+    num_urls = len(app.state.config.OPENAI_API_BASE_URLS)
+    num_keys = len(app.state.config.OPENAI_API_KEYS)
+
+    if num_keys != num_urls:
+        # if there are more keys than urls, remove the extra keys
+        if num_keys > num_urls:
+            new_keys = app.state.config.OPENAI_API_KEYS[:num_urls]
+            app.state.config.OPENAI_API_KEYS = new_keys
+        # if there are more urls than keys, add empty keys
+        else:
+            app.state.config.OPENAI_API_KEYS += [""] * (num_urls - num_keys)
+
+    tasks = [
+        fetch_url(f"{url}/models", app.state.config.OPENAI_API_KEYS[idx])
+        for idx, url in enumerate(app.state.config.OPENAI_API_BASE_URLS)
+    ]
+
+    responses = await asyncio.gather(*tasks)
+    log.debug(f"get_all_models:responses() {responses}")
+
+    return responses
+
+
+@overload
+async def get_all_models(raw: Literal[True]) -> list: ...
+
+
+@overload
+async def get_all_models(raw: Literal[False] = False) -> dict[str, list]: ...
+
+
+async def get_all_models(raw=False) -> dict[str, list] | list:
     log.info("get_all_models()")
+    if is_openai_api_disabled():
+        return [] if raw else {"data": []}
 
-    if (
-        len(app.state.config.OPENAI_API_KEYS) == 1
-        and app.state.config.OPENAI_API_KEYS[0] == ""
-    ) or not app.state.config.ENABLE_OPENAI_API:
-        models = {"data": []}
-    else:
-        # Check if API KEYS length is same than API URLS length
-        if len(app.state.config.OPENAI_API_KEYS) != len(
-            app.state.config.OPENAI_API_BASE_URLS
-        ):
-            # if there are more keys than urls, remove the extra keys
-            if len(app.state.config.OPENAI_API_KEYS) > len(
-                app.state.config.OPENAI_API_BASE_URLS
-            ):
-                app.state.config.OPENAI_API_KEYS = app.state.config.OPENAI_API_KEYS[
-                    : len(app.state.config.OPENAI_API_BASE_URLS)
-                ]
-            # if there are more urls than keys, add empty keys
-            else:
-                app.state.config.OPENAI_API_KEYS += [
-                    ""
-                    for _ in range(
-                        len(app.state.config.OPENAI_API_BASE_URLS)
-                        - len(app.state.config.OPENAI_API_KEYS)
-                    )
-                ]
+    responses = await get_all_models_raw()
+    if raw:
+        return responses
 
-        tasks = [
-            fetch_url(f"{url}/models", app.state.config.OPENAI_API_KEYS[idx])
-            for idx, url in enumerate(app.state.config.OPENAI_API_BASE_URLS)
-        ]
+    def extract_data(response):
+        if response and "data" in response:
+            return response["data"]
+        if isinstance(response, list):
+            return response
+        return None
 
-        responses = await asyncio.gather(*tasks)
-        log.debug(f"get_all_models:responses() {responses}")
+    models = {"data": merge_models_lists(map(extract_data, responses))}
 
-        if raw:
-            return responses
-
-        models = {
-            "data": merge_models_lists(
-                list(
-                    map(
-                        lambda response: (
-                            response["data"]
-                            if (response and "data" in response)
-                            else (response if isinstance(response, list) else None)
-                        ),
-                        responses,
-                    )
-                )
-            )
-        }
-
-        log.debug(f"models: {models}")
-        app.state.MODELS = {model["id"]: model for model in models["data"]}
+    log.debug(f"models: {models}")
+    app.state.MODELS = {model["id"]: model for model in models["data"]}
 
     return models
 
@@ -297,7 +306,7 @@ async def get_all_models(raw: bool = False):
 @app.get("/models")
 @app.get("/models/{url_idx}")
 async def get_models(url_idx: Optional[int] = None, user=Depends(get_verified_user)):
-    if url_idx == None:
+    if url_idx is None:
         models = await get_all_models()
         if app.state.config.ENABLE_MODEL_FILTER:
             if user.role == "user":
@@ -324,10 +333,24 @@ async def get_models(url_idx: Optional[int] = None, user=Depends(get_verified_us
             r.raise_for_status()
 
             response_data = r.json()
+
             if "api.openai.com" in url:
-                response_data["data"] = list(
-                    filter(lambda model: "gpt" in model["id"], response_data["data"])
-                )
+                # Filter the response data
+                response_data["data"] = [
+                    model
+                    for model in response_data["data"]
+                    if not any(
+                        name in model["id"]
+                        for name in [
+                            "babbage",
+                            "dall-e",
+                            "davinci",
+                            "embedding",
+                            "tts",
+                            "whisper",
+                        ]
+                    )
+                ]
 
             return response_data
         except Exception as e:
@@ -338,7 +361,7 @@ async def get_models(url_idx: Optional[int] = None, user=Depends(get_verified_us
                     res = r.json()
                     if "error" in res:
                         error_detail = f"External: {res['error']}"
-                except:
+                except Exception:
                     error_detail = f"External: {e}"
 
             raise HTTPException(
@@ -357,6 +380,9 @@ async def generate_chat_completion(
     idx = 0
     payload = {**form_data}
 
+    if "metadata" in payload:
+        del payload["metadata"]
+
     model_id = form_data.get("model")
     model_info = Models.get_model_by_id(model_id)
 
@@ -364,69 +390,9 @@ async def generate_chat_completion(
         if model_info.base_model_id:
             payload["model"] = model_info.base_model_id
 
-        model_info.params = model_info.params.model_dump()
-
-        if model_info.params:
-            if model_info.params.get("temperature", None) is not None:
-                payload["temperature"] = float(model_info.params.get("temperature"))
-
-            if model_info.params.get("top_p", None):
-                payload["top_p"] = int(model_info.params.get("top_p", None))
-
-            if model_info.params.get("max_tokens", None):
-                payload["max_tokens"] = int(model_info.params.get("max_tokens", None))
-
-            if model_info.params.get("frequency_penalty", None):
-                payload["frequency_penalty"] = int(
-                    model_info.params.get("frequency_penalty", None)
-                )
-
-            if model_info.params.get("seed", None):
-                payload["seed"] = model_info.params.get("seed", None)
-
-            if model_info.params.get("stop", None):
-                payload["stop"] = (
-                    [
-                        bytes(stop, "utf-8").decode("unicode_escape")
-                        for stop in model_info.params["stop"]
-                    ]
-                    if model_info.params.get("stop", None)
-                    else None
-                )
-
-        system = model_info.params.get("system", None)
-        if system:
-            system = prompt_template(
-                system,
-                **(
-                    {
-                        "user_name": user.name,
-                        "user_location": (
-                            user.info.get("location") if user.info else None
-                        ),
-                    }
-                    if user
-                    else {}
-                ),
-            )
-            # Check if the payload already has a system message
-            # If not, add a system message to the payload
-            if payload.get("messages"):
-                for message in payload["messages"]:
-                    if message.get("role") == "system":
-                        message["content"] = system + message["content"]
-                        break
-                else:
-                    payload["messages"].insert(
-                        0,
-                        {
-                            "role": "system",
-                            "content": system,
-                        },
-                    )
-
-    else:
-        pass
+        params = model_info.params.model_dump()
+        payload = apply_model_params_to_body_openai(params, payload)
+        payload = apply_model_system_prompt_to_body(params, payload, user)
 
     model = app.state.MODELS[payload.get("model")]
     idx = model["urlIdx"]
@@ -439,31 +405,40 @@ async def generate_chat_completion(
             "role": user.role,
         }
 
-    # Check if the model is "gpt-4-vision-preview" and set "max_tokens" to 4000
-    # This is a workaround until OpenAI fixes the issue with this model
-    if payload.get("model") == "gpt-4-vision-preview":
-        if "max_tokens" not in payload:
-            payload["max_tokens"] = 4000
-        log.debug("Modified payload:", payload)
+    url = app.state.config.OPENAI_API_BASE_URLS[idx]
+    key = app.state.config.OPENAI_API_KEYS[idx]
+
+    # Change max_completion_tokens to max_tokens (Backward compatible)
+    if "api.openai.com" not in url and not payload["model"].lower().startswith("o1-"):
+        if "max_completion_tokens" in payload:
+            # Remove "max_completion_tokens" from the payload
+            payload["max_tokens"] = payload["max_completion_tokens"]
+            del payload["max_completion_tokens"]
+    else:
+        if "max_tokens" in payload and "max_completion_tokens" in payload:
+            del payload["max_tokens"]
 
     # Convert the modified body back to JSON
     payload = json.dumps(payload)
 
     log.debug(payload)
 
-    url = app.state.config.OPENAI_API_BASE_URLS[idx]
-    key = app.state.config.OPENAI_API_KEYS[idx]
-
     headers = {}
     headers["Authorization"] = f"Bearer {key}"
     headers["Content-Type"] = "application/json"
+    if "openrouter.ai" in app.state.config.OPENAI_API_BASE_URLS[idx]:
+        headers["HTTP-Referer"] = "https://openwebui.com/"
+        headers["X-Title"] = "Open WebUI"
 
     r = None
     session = None
     streaming = False
+    response = None
 
     try:
-        session = aiohttp.ClientSession(trust_env=True)
+        session = aiohttp.ClientSession(
+            trust_env=True, timeout=aiohttp.ClientTimeout(total=AIOHTTP_CLIENT_TIMEOUT)
+        )
         r = await session.request(
             method="POST",
             url=f"{url}/chat/completions",
@@ -471,8 +446,6 @@ async def generate_chat_completion(
             headers=headers,
         )
 
-        r.raise_for_status()
-
         # Check if response is SSE
         if "text/event-stream" in r.headers.get("Content-Type", ""):
             streaming = True
@@ -485,19 +458,23 @@ async def generate_chat_completion(
                 ),
             )
         else:
-            response_data = await r.json()
-            return response_data
+            try:
+                response = await r.json()
+            except Exception as e:
+                log.error(e)
+                response = await r.text()
+
+            r.raise_for_status()
+            return response
     except Exception as e:
         log.exception(e)
         error_detail = "Open WebUI: Server Connection Error"
-        if r is not None:
-            try:
-                res = await r.json()
-                print(res)
-                if "error" in res:
-                    error_detail = f"External: {res['error']['message'] if 'message' in res['error'] else res['error']}"
-            except:
-                error_detail = f"External: {e}"
+        if isinstance(response, dict):
+            if "error" in response:
+                error_detail = f"{response['error']['message'] if 'message' in response['error'] else response['error']}"
+        elif isinstance(response, str):
+            error_detail = response
+
         raise HTTPException(status_code=r.status if r else 500, detail=error_detail)
     finally:
         if not streaming and session:
@@ -559,7 +536,7 @@ async def proxy(path: str, request: Request, user=Depends(get_verified_user)):
                 print(res)
                 if "error" in res:
                     error_detail = f"External: {res['error']['message'] if 'message' in res['error'] else res['error']}"
-            except:
+            except Exception:
                 error_detail = f"External: {e}"
         raise HTTPException(status_code=r.status if r else 500, detail=error_detail)
     finally:

backend/open_webui/apps/rag/main.py

@@ -1,137 +1,125 @@
-from fastapi import (
-    FastAPI,
-    Depends,
-    HTTPException,
-    status,
-    UploadFile,
-    File,
-    Form,
-)
-from fastapi.middleware.cors import CORSMiddleware
-import requests
-import os, shutil, logging, re
-from datetime import datetime
-
-from pathlib import Path
-from typing import List, Union, Sequence, Iterator, Any
-
-from chromadb.utils.batch_utils import create_batches
-from langchain_core.documents import Document
-
-from langchain_community.document_loaders import (
-    WebBaseLoader,
-    TextLoader,
-    PyPDFLoader,
-    CSVLoader,
-    BSHTMLLoader,
-    Docx2txtLoader,
-    UnstructuredEPubLoader,
-    UnstructuredWordDocumentLoader,
-    UnstructuredMarkdownLoader,
-    UnstructuredXMLLoader,
-    UnstructuredRSTLoader,
-    UnstructuredExcelLoader,
-    UnstructuredPowerPointLoader,
-    YoutubeLoader,
-    OutlookMessageLoader,
-)
-from langchain.text_splitter import RecursiveCharacterTextSplitter
-
-import validators
-import urllib.parse
-import socket
-
-
-from pydantic import BaseModel
-from typing import Optional
-import mimetypes
-import uuid
 import json
+import logging
+import mimetypes
+import os
+import shutil
+import socket
+import urllib.parse
+import uuid
+from datetime import datetime
+from pathlib import Path
+from typing import Iterator, Optional, Sequence, Union
 
-import sentence_transformers
 
-from apps.webui.models.documents import (
-    Documents,
-    DocumentForm,
-    DocumentResponse,
-)
-from apps.webui.models.files import (
-    Files,
-)
+import numpy as np
+import torch
+import requests
+import validators
 
-from apps.rag.utils import (
-    get_model_path,
+from fastapi import Depends, FastAPI, File, Form, HTTPException, UploadFile, status
+from fastapi.middleware.cors import CORSMiddleware
+from pydantic import BaseModel
+
+from open_webui.apps.rag.search.main import SearchResult
+from open_webui.apps.rag.search.brave import search_brave
+from open_webui.apps.rag.search.duckduckgo import search_duckduckgo
+from open_webui.apps.rag.search.google_pse import search_google_pse
+from open_webui.apps.rag.search.jina_search import search_jina
+from open_webui.apps.rag.search.searchapi import search_searchapi
+from open_webui.apps.rag.search.searxng import search_searxng
+from open_webui.apps.rag.search.serper import search_serper
+from open_webui.apps.rag.search.serply import search_serply
+from open_webui.apps.rag.search.serpstack import search_serpstack
+from open_webui.apps.rag.search.tavily import search_tavily
+from open_webui.apps.rag.utils import (
     get_embedding_function,
-    query_doc,
-    query_doc_with_hybrid_search,
+    get_model_path,
     query_collection,
     query_collection_with_hybrid_search,
+    query_doc,
+    query_doc_with_hybrid_search,
 )
-
-from apps.rag.search.brave import search_brave
-from apps.rag.search.google_pse import search_google_pse
-from apps.rag.search.main import SearchResult
-from apps.rag.search.searxng import search_searxng
-from apps.rag.search.serper import search_serper
-from apps.rag.search.serpstack import search_serpstack
-from apps.rag.search.serply import search_serply
-from apps.rag.search.duckduckgo import search_duckduckgo
-from apps.rag.search.tavily import search_tavily
-from apps.rag.search.jina_search import search_jina
-
-from utils.misc import (
-    calculate_sha256,
-    calculate_sha256_string,
-    sanitize_filename,
-    extract_folders_after_data_docs,
-)
-from utils.utils import get_verified_user, get_admin_user
-
-from config import (
-    AppConfig,
-    ENV,
-    SRC_LOG_LEVELS,
-    UPLOAD_DIR,
+from open_webui.apps.webui.models.documents import DocumentForm, Documents
+from open_webui.apps.webui.models.files import Files
+from open_webui.config import (
+    BRAVE_SEARCH_API_KEY,
+    CHUNK_OVERLAP,
+    CHUNK_SIZE,
+    CONTENT_EXTRACTION_ENGINE,
+    CORS_ALLOW_ORIGIN,
     DOCS_DIR,
-    RAG_TOP_K,
-    RAG_RELEVANCE_THRESHOLD,
+    ENABLE_RAG_HYBRID_SEARCH,
+    ENABLE_RAG_LOCAL_WEB_FETCH,
+    ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
+    ENABLE_RAG_WEB_SEARCH,
+    ENV,
+    GOOGLE_PSE_API_KEY,
+    GOOGLE_PSE_ENGINE_ID,
+    PDF_EXTRACT_IMAGES,
     RAG_EMBEDDING_ENGINE,
     RAG_EMBEDDING_MODEL,
     RAG_EMBEDDING_MODEL_AUTO_UPDATE,
     RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE,
-    ENABLE_RAG_HYBRID_SEARCH,
-    ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION,
-    RAG_RERANKING_MODEL,
-    PDF_EXTRACT_IMAGES,
-    RAG_RERANKING_MODEL_AUTO_UPDATE,
-    RAG_RERANKING_MODEL_TRUST_REMOTE_CODE,
+    RAG_EMBEDDING_OPENAI_BATCH_SIZE,
+    RAG_FILE_MAX_COUNT,
+    RAG_FILE_MAX_SIZE,
     RAG_OPENAI_API_BASE_URL,
     RAG_OPENAI_API_KEY,
-    DEVICE_TYPE,
-    CHROMA_CLIENT,
-    CHUNK_SIZE,
-    CHUNK_OVERLAP,
+    RAG_RELEVANCE_THRESHOLD,
+    RAG_RERANKING_MODEL,
+    RAG_RERANKING_MODEL_AUTO_UPDATE,
+    RAG_RERANKING_MODEL_TRUST_REMOTE_CODE,
+    DEFAULT_RAG_TEMPLATE,
     RAG_TEMPLATE,
-    ENABLE_RAG_LOCAL_WEB_FETCH,
-    YOUTUBE_LOADER_LANGUAGE,
-    ENABLE_RAG_WEB_SEARCH,
-    RAG_WEB_SEARCH_ENGINE,
+    RAG_TOP_K,
+    RAG_WEB_SEARCH_CONCURRENT_REQUESTS,
     RAG_WEB_SEARCH_DOMAIN_FILTER_LIST,
+    RAG_WEB_SEARCH_ENGINE,
+    RAG_WEB_SEARCH_RESULT_COUNT,
+    SEARCHAPI_API_KEY,
+    SEARCHAPI_ENGINE,
     SEARXNG_QUERY_URL,
-    GOOGLE_PSE_API_KEY,
-    GOOGLE_PSE_ENGINE_ID,
-    BRAVE_SEARCH_API_KEY,
-    SERPSTACK_API_KEY,
-    SERPSTACK_HTTPS,
     SERPER_API_KEY,
     SERPLY_API_KEY,
+    SERPSTACK_API_KEY,
+    SERPSTACK_HTTPS,
     TAVILY_API_KEY,
-    RAG_WEB_SEARCH_RESULT_COUNT,
-    RAG_WEB_SEARCH_CONCURRENT_REQUESTS,
-    RAG_EMBEDDING_OPENAI_BATCH_SIZE,
+    TIKA_SERVER_URL,
+    UPLOAD_DIR,
+    YOUTUBE_LOADER_LANGUAGE,
+    AppConfig,
 )
+from open_webui.constants import ERROR_MESSAGES
+from open_webui.env import SRC_LOG_LEVELS, DEVICE_TYPE, DOCKER
+from open_webui.utils.misc import (
+    calculate_sha256,
+    calculate_sha256_string,
+    extract_folders_after_data_docs,
+    sanitize_filename,
+)
+from open_webui.utils.utils import get_admin_user, get_verified_user
+from open_webui.apps.rag.vector.connector import VECTOR_DB_CLIENT
 
-from constants import ERROR_MESSAGES
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.document_loaders import (
+    BSHTMLLoader,
+    CSVLoader,
+    Docx2txtLoader,
+    OutlookMessageLoader,
+    PyPDFLoader,
+    TextLoader,
+    UnstructuredEPubLoader,
+    UnstructuredExcelLoader,
+    UnstructuredMarkdownLoader,
+    UnstructuredPowerPointLoader,
+    UnstructuredRSTLoader,
+    UnstructuredXMLLoader,
+    WebBaseLoader,
+    YoutubeLoader,
+)
+from langchain_core.documents import Document
+from colbert.infra import ColBERTConfig
+from colbert.modeling.checkpoint import Checkpoint
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
@@ -142,12 +130,17 @@ app.state.config = AppConfig()
 
 app.state.config.TOP_K = RAG_TOP_K
 app.state.config.RELEVANCE_THRESHOLD = RAG_RELEVANCE_THRESHOLD
+app.state.config.FILE_MAX_SIZE = RAG_FILE_MAX_SIZE
+app.state.config.FILE_MAX_COUNT = RAG_FILE_MAX_COUNT
 
 app.state.config.ENABLE_RAG_HYBRID_SEARCH = ENABLE_RAG_HYBRID_SEARCH
 app.state.config.ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION = (
     ENABLE_RAG_WEB_LOADER_SSL_VERIFICATION
 )
 
+app.state.config.CONTENT_EXTRACTION_ENGINE = CONTENT_EXTRACTION_ENGINE
+app.state.config.TIKA_SERVER_URL = TIKA_SERVER_URL
+
 app.state.config.CHUNK_SIZE = CHUNK_SIZE
 app.state.config.CHUNK_OVERLAP = CHUNK_OVERLAP
 
@@ -157,13 +150,11 @@ app.state.config.RAG_EMBEDDING_OPENAI_BATCH_SIZE = RAG_EMBEDDING_OPENAI_BATCH_SI
 app.state.config.RAG_RERANKING_MODEL = RAG_RERANKING_MODEL
 app.state.config.RAG_TEMPLATE = RAG_TEMPLATE
 
-
 app.state.config.OPENAI_API_BASE_URL = RAG_OPENAI_API_BASE_URL
 app.state.config.OPENAI_API_KEY = RAG_OPENAI_API_KEY
 
 app.state.config.PDF_EXTRACT_IMAGES = PDF_EXTRACT_IMAGES
 
-
 app.state.config.YOUTUBE_LOADER_LANGUAGE = YOUTUBE_LOADER_LANGUAGE
 app.state.YOUTUBE_LOADER_TRANSLATION = None
 
@@ -181,17 +172,21 @@ app.state.config.SERPSTACK_HTTPS = SERPSTACK_HTTPS
 app.state.config.SERPER_API_KEY = SERPER_API_KEY
 app.state.config.SERPLY_API_KEY = SERPLY_API_KEY
 app.state.config.TAVILY_API_KEY = TAVILY_API_KEY
+app.state.config.SEARCHAPI_API_KEY = SEARCHAPI_API_KEY
+app.state.config.SEARCHAPI_ENGINE = SEARCHAPI_ENGINE
 app.state.config.RAG_WEB_SEARCH_RESULT_COUNT = RAG_WEB_SEARCH_RESULT_COUNT
 app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS = RAG_WEB_SEARCH_CONCURRENT_REQUESTS
 
 
 def update_embedding_model(
     embedding_model: str,
-    update_model: bool = False,
+    auto_update: bool = False,
 ):
     if embedding_model and app.state.config.RAG_EMBEDDING_ENGINE == "":
+        import sentence_transformers
+
         app.state.sentence_transformer_ef = sentence_transformers.SentenceTransformer(
-            get_model_path(embedding_model, update_model),
+            get_model_path(embedding_model, auto_update),
             device=DEVICE_TYPE,
             trust_remote_code=RAG_EMBEDDING_MODEL_TRUST_REMOTE_CODE,
         )
@@ -201,14 +196,108 @@ def update_embedding_model(
 
 def update_reranking_model(
     reranking_model: str,
-    update_model: bool = False,
+    auto_update: bool = False,
 ):
     if reranking_model:
-        app.state.sentence_transformer_rf = sentence_transformers.CrossEncoder(
-            get_model_path(reranking_model, update_model),
-            device=DEVICE_TYPE,
-            trust_remote_code=RAG_RERANKING_MODEL_TRUST_REMOTE_CODE,
-        )
+        if any(model in reranking_model for model in ["jinaai/jina-colbert-v2"]):
+
+            class ColBERT:
+                def __init__(self, name) -> None:
+                    print("ColBERT: Loading model", name)
+                    self.device = "cuda" if torch.cuda.is_available() else "cpu"
+
+                    if DOCKER:
+                        # This is a workaround for the issue with the docker container
+                        # where the torch extension is not loaded properly
+                        # and the following error is thrown:
+                        # /root/.cache/torch_extensions/py311_cpu/segmented_maxsim_cpp/segmented_maxsim_cpp.so: cannot open shared object file: No such file or directory
+
+                        lock_file = "/root/.cache/torch_extensions/py311_cpu/segmented_maxsim_cpp/lock"
+                        if os.path.exists(lock_file):
+                            os.remove(lock_file)
+
+                    self.ckpt = Checkpoint(
+                        name,
+                        colbert_config=ColBERTConfig(model_name=name),
+                    ).to(self.device)
+                    pass
+
+                def calculate_similarity_scores(
+                    self, query_embeddings, document_embeddings
+                ):
+
+                    query_embeddings = query_embeddings.to(self.device)
+                    document_embeddings = document_embeddings.to(self.device)
+
+                    # Validate dimensions to ensure compatibility
+                    if query_embeddings.dim() != 3:
+                        raise ValueError(
+                            f"Expected query embeddings to have 3 dimensions, but got {query_embeddings.dim()}."
+                        )
+                    if document_embeddings.dim() != 3:
+                        raise ValueError(
+                            f"Expected document embeddings to have 3 dimensions, but got {document_embeddings.dim()}."
+                        )
+                    if query_embeddings.size(0) not in [1, document_embeddings.size(0)]:
+                        raise ValueError(
+                            "There should be either one query or queries equal to the number of documents."
+                        )
+
+                    # Transpose the query embeddings to align for matrix multiplication
+                    transposed_query_embeddings = query_embeddings.permute(0, 2, 1)
+                    # Compute similarity scores using batch matrix multiplication
+                    computed_scores = torch.matmul(
+                        document_embeddings, transposed_query_embeddings
+                    )
+                    # Apply max pooling to extract the highest semantic similarity across each document's sequence
+                    maximum_scores = torch.max(computed_scores, dim=1).values
+
+                    # Sum up the maximum scores across features to get the overall document relevance scores
+                    final_scores = maximum_scores.sum(dim=1)
+
+                    normalized_scores = torch.softmax(final_scores, dim=0)
+
+                    return normalized_scores.detach().cpu().numpy().astype(np.float32)
+
+                def predict(self, sentences):
+
+                    query = sentences[0][0]
+                    docs = [i[1] for i in sentences]
+
+                    # Embedding the documents
+                    embedded_docs = self.ckpt.docFromText(docs, bsize=32)[0]
+                    # Embedding the queries
+                    embedded_queries = self.ckpt.queryFromText([query], bsize=32)
+                    embedded_query = embedded_queries[0]
+
+                    # Calculate retrieval scores for the query against all documents
+                    scores = self.calculate_similarity_scores(
+                        embedded_query.unsqueeze(0), embedded_docs
+                    )
+
+                    return scores
+
+            try:
+                app.state.sentence_transformer_rf = ColBERT(
+                    get_model_path(reranking_model, auto_update)
+                )
+            except Exception as e:
+                log.error(f"ColBERT: {e}")
+                app.state.sentence_transformer_rf = None
+                app.state.config.ENABLE_RAG_HYBRID_SEARCH = False
+        else:
+            import sentence_transformers
+
+            try:
+                app.state.sentence_transformer_rf = sentence_transformers.CrossEncoder(
+                    get_model_path(reranking_model, auto_update),
+                    device=DEVICE_TYPE,
+                    trust_remote_code=RAG_RERANKING_MODEL_TRUST_REMOTE_CODE,
+                )
+            except:
+                log.error("CrossEncoder error")
+                app.state.sentence_transformer_rf = None
+                app.state.config.ENABLE_RAG_HYBRID_SEARCH = False
     else:
         app.state.sentence_transformer_rf = None
 
@@ -233,12 +322,9 @@ app.state.EMBEDDING_FUNCTION = get_embedding_function(
     app.state.config.RAG_EMBEDDING_OPENAI_BATCH_SIZE,
 )
 
-origins = ["*"]
-
-
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=origins,
+    allow_origins=CORS_ALLOW_ORIGIN,
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],
@@ -369,7 +455,7 @@ async def update_reranking_config(
     try:
         app.state.config.RAG_RERANKING_MODEL = form_data.reranking_model
 
-        update_reranking_model(app.state.config.RAG_RERANKING_MODEL), True
+        update_reranking_model(app.state.config.RAG_RERANKING_MODEL, True)
 
         return {
             "status": True,
@@ -388,6 +474,14 @@ async def get_rag_config(user=Depends(get_admin_user)):
     return {
         "status": True,
         "pdf_extract_images": app.state.config.PDF_EXTRACT_IMAGES,
+        "file": {
+            "max_size": app.state.config.FILE_MAX_SIZE,
+            "max_count": app.state.config.FILE_MAX_COUNT,
+        },
+        "content_extraction": {
+            "engine": app.state.config.CONTENT_EXTRACTION_ENGINE,
+            "tika_server_url": app.state.config.TIKA_SERVER_URL,
+        },
         "chunk": {
             "chunk_size": app.state.config.CHUNK_SIZE,
             "chunk_overlap": app.state.config.CHUNK_OVERLAP,
@@ -410,6 +504,8 @@ async def get_rag_config(user=Depends(get_admin_user)):
                 "serper_api_key": app.state.config.SERPER_API_KEY,
                 "serply_api_key": app.state.config.SERPLY_API_KEY,
                 "tavily_api_key": app.state.config.TAVILY_API_KEY,
+                "searchapi_api_key": app.state.config.SEARCHAPI_API_KEY,
+                "seaarchapi_engine": app.state.config.SEARCHAPI_ENGINE,
                 "result_count": app.state.config.RAG_WEB_SEARCH_RESULT_COUNT,
                 "concurrent_requests": app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS,
             },
@@ -417,13 +513,23 @@ async def get_rag_config(user=Depends(get_admin_user)):
     }
 
 
+class FileConfig(BaseModel):
+    max_size: Optional[int] = None
+    max_count: Optional[int] = None
+
+
+class ContentExtractionConfig(BaseModel):
+    engine: str = ""
+    tika_server_url: Optional[str] = None
+
+
 class ChunkParamUpdateForm(BaseModel):
     chunk_size: int
     chunk_overlap: int
 
 
 class YoutubeLoaderConfig(BaseModel):
-    language: List[str]
+    language: list[str]
     translation: Optional[str] = None
 
 
@@ -439,6 +545,8 @@ class WebSearchConfig(BaseModel):
     serper_api_key: Optional[str] = None
     serply_api_key: Optional[str] = None
     tavily_api_key: Optional[str] = None
+    searchapi_api_key: Optional[str] = None
+    searchapi_engine: Optional[str] = None
     result_count: Optional[int] = None
     concurrent_requests: Optional[int] = None
 
@@ -450,6 +558,8 @@ class WebConfig(BaseModel):
 
 class ConfigUpdateForm(BaseModel):
     pdf_extract_images: Optional[bool] = None
+    file: Optional[FileConfig] = None
+    content_extraction: Optional[ContentExtractionConfig] = None
     chunk: Optional[ChunkParamUpdateForm] = None
     youtube: Optional[YoutubeLoaderConfig] = None
     web: Optional[WebConfig] = None
@@ -463,6 +573,15 @@ async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_
         else app.state.config.PDF_EXTRACT_IMAGES
     )
 
+    if form_data.file is not None:
+        app.state.config.FILE_MAX_SIZE = form_data.file.max_size
+        app.state.config.FILE_MAX_COUNT = form_data.file.max_count
+
+    if form_data.content_extraction is not None:
+        log.info(f"Updating text settings: {form_data.content_extraction}")
+        app.state.config.CONTENT_EXTRACTION_ENGINE = form_data.content_extraction.engine
+        app.state.config.TIKA_SERVER_URL = form_data.content_extraction.tika_server_url
+
     if form_data.chunk is not None:
         app.state.config.CHUNK_SIZE = form_data.chunk.chunk_size
         app.state.config.CHUNK_OVERLAP = form_data.chunk.chunk_overlap
@@ -491,6 +610,8 @@ async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_
         app.state.config.SERPER_API_KEY = form_data.web.search.serper_api_key
         app.state.config.SERPLY_API_KEY = form_data.web.search.serply_api_key
         app.state.config.TAVILY_API_KEY = form_data.web.search.tavily_api_key
+        app.state.config.SEARCHAPI_API_KEY = form_data.web.search.searchapi_api_key
+        app.state.config.SEARCHAPI_ENGINE = form_data.web.search.searchapi_engine
         app.state.config.RAG_WEB_SEARCH_RESULT_COUNT = form_data.web.search.result_count
         app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS = (
             form_data.web.search.concurrent_requests
@@ -499,6 +620,14 @@ async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_
     return {
         "status": True,
         "pdf_extract_images": app.state.config.PDF_EXTRACT_IMAGES,
+        "file": {
+            "max_size": app.state.config.FILE_MAX_SIZE,
+            "max_count": app.state.config.FILE_MAX_COUNT,
+        },
+        "content_extraction": {
+            "engine": app.state.config.CONTENT_EXTRACTION_ENGINE,
+            "tika_server_url": app.state.config.TIKA_SERVER_URL,
+        },
         "chunk": {
             "chunk_size": app.state.config.CHUNK_SIZE,
             "chunk_overlap": app.state.config.CHUNK_OVERLAP,
@@ -520,6 +649,8 @@ async def update_rag_config(form_data: ConfigUpdateForm, user=Depends(get_admin_
                 "serpstack_https": app.state.config.SERPSTACK_HTTPS,
                 "serper_api_key": app.state.config.SERPER_API_KEY,
                 "serply_api_key": app.state.config.SERPLY_API_KEY,
+                "serachapi_api_key": app.state.config.SEARCHAPI_API_KEY,
+                "searchapi_engine": app.state.config.SEARCHAPI_ENGINE,
                 "tavily_api_key": app.state.config.TAVILY_API_KEY,
                 "result_count": app.state.config.RAG_WEB_SEARCH_RESULT_COUNT,
                 "concurrent_requests": app.state.config.RAG_WEB_SEARCH_CONCURRENT_REQUESTS,
@@ -559,13 +690,14 @@ async def update_query_settings(
     form_data: QuerySettingsForm, user=Depends(get_admin_user)
 ):
     app.state.config.RAG_TEMPLATE = (
-        form_data.template if form_data.template else RAG_TEMPLATE
+        form_data.template if form_data.template != "" else DEFAULT_RAG_TEMPLATE
     )
     app.state.config.TOP_K = form_data.k if form_data.k else 4
     app.state.config.RELEVANCE_THRESHOLD = form_data.r if form_data.r else 0.0
     app.state.config.ENABLE_RAG_HYBRID_SEARCH = (
         form_data.hybrid if form_data.hybrid else False
     )
+
     return {
         "status": True,
         "template": app.state.config.RAG_TEMPLATE,
@@ -616,7 +748,7 @@ def query_doc_handler(
 
 
 class QueryCollectionsForm(BaseModel):
-    collection_names: List[str]
+    collection_names: list[str]
     query: str
     k: Optional[int] = None
     r: Optional[float] = None
@@ -770,6 +902,7 @@ def search_web(engine: str, query: str) -> list[SearchResult]:
     - SERPER_API_KEY
     - SERPLY_API_KEY
     - TAVILY_API_KEY
+    - SEARCHAPI_API_KEY + SEARCHAPI_ENGINE (by default `google`)
     Args:
         query (str): The query to search for
     """
@@ -857,6 +990,17 @@ def search_web(engine: str, query: str) -> list[SearchResult]:
             )
         else:
             raise Exception("No TAVILY_API_KEY found in environment variables")
+    elif engine == "searchapi":
+        if app.state.config.SEARCHAPI_API_KEY:
+            return search_searchapi(
+                app.state.config.SEARCHAPI_API_KEY,
+                app.state.config.SEARCHAPI_ENGINE,
+                query,
+                app.state.config.RAG_WEB_SEARCH_RESULT_COUNT,
+                app.state.config.RAG_WEB_SEARCH_DOMAIN_FILTER_LIST,
+            )
+        else:
+            raise Exception("No SEARCHAPI_API_KEY found in environment variables")
     elif engine == "jina":
         return search_jina(query, app.state.config.RAG_WEB_SEARCH_RESULT_COUNT)
     else:
@@ -904,8 +1048,9 @@ def store_web_search(form_data: SearchForm, user=Depends(get_verified_user)):
         )
 
 
-def store_data_in_vector_db(data, collection_name, overwrite: bool = False) -> bool:
-
+def store_data_in_vector_db(
+    data, collection_name, metadata: Optional[dict] = None, overwrite: bool = False
+) -> bool:
     text_splitter = RecursiveCharacterTextSplitter(
         chunk_size=app.state.config.CHUNK_SIZE,
         chunk_overlap=app.state.config.CHUNK_OVERLAP,
@@ -916,7 +1061,7 @@ def store_data_in_vector_db(data, collection_name, overwrite: bool = False) -> b
 
     if len(docs) > 0:
         log.info(f"store_data_in_vector_db {docs}")
-        return store_docs_in_vector_db(docs, collection_name, overwrite), None
+        return store_docs_in_vector_db(docs, collection_name, metadata, overwrite), None
     else:
         raise ValueError(ERROR_MESSAGES.EMPTY_CONTENT)
 
@@ -930,14 +1075,16 @@ def store_text_in_vector_db(
         add_start_index=True,
     )
     docs = text_splitter.create_documents([text], metadatas=[metadata])
-    return store_docs_in_vector_db(docs, collection_name, overwrite)
+    return store_docs_in_vector_db(docs, collection_name, overwrite=overwrite)
 
 
-def store_docs_in_vector_db(docs, collection_name, overwrite: bool = False) -> bool:
+def store_docs_in_vector_db(
+    docs, collection_name, metadata: Optional[dict] = None, overwrite: bool = False
+) -> bool:
     log.info(f"store_docs_in_vector_db {docs} {collection_name}")
 
     texts = [doc.page_content for doc in docs]
-    metadatas = [doc.metadata for doc in docs]
+    metadatas = [{**doc.metadata, **(metadata if metadata else {})} for doc in docs]
 
     # ChromaDB does not like datetime formats
     # for meta-data so convert them to string.
@@ -948,43 +1095,77 @@ def store_docs_in_vector_db(docs, collection_name, overwrite: bool = False) -> b
 
     try:
         if overwrite:
-            for collection in CHROMA_CLIENT.list_collections():
-                if collection_name == collection.name:
-                    log.info(f"deleting existing collection {collection_name}")
-                    CHROMA_CLIENT.delete_collection(name=collection_name)
+            if VECTOR_DB_CLIENT.has_collection(collection_name=collection_name):
+                log.info(f"deleting existing collection {collection_name}")
+                VECTOR_DB_CLIENT.delete_collection(collection_name=collection_name)
 
-        collection = CHROMA_CLIENT.create_collection(name=collection_name)
+        if VECTOR_DB_CLIENT.has_collection(collection_name=collection_name):
+            log.info(f"collection {collection_name} already exists")
+            return True
+        else:
+            embedding_function = get_embedding_function(
+                app.state.config.RAG_EMBEDDING_ENGINE,
+                app.state.config.RAG_EMBEDDING_MODEL,
+                app.state.sentence_transformer_ef,
+                app.state.config.OPENAI_API_KEY,
+                app.state.config.OPENAI_API_BASE_URL,
+                app.state.config.RAG_EMBEDDING_OPENAI_BATCH_SIZE,
+            )
 
-        embedding_func = get_embedding_function(
-            app.state.config.RAG_EMBEDDING_ENGINE,
-            app.state.config.RAG_EMBEDDING_MODEL,
-            app.state.sentence_transformer_ef,
-            app.state.config.OPENAI_API_KEY,
-            app.state.config.OPENAI_API_BASE_URL,
-            app.state.config.RAG_EMBEDDING_OPENAI_BATCH_SIZE,
-        )
+            VECTOR_DB_CLIENT.insert(
+                collection_name=collection_name,
+                items=[
+                    {
+                        "id": str(uuid.uuid4()),
+                        "text": text,
+                        "vector": embedding_function(text.replace("\n", " ")),
+                        "metadata": metadatas[idx],
+                    }
+                    for idx, text in enumerate(texts)
+                ],
+            )
 
-        embedding_texts = list(map(lambda x: x.replace("\n", " "), texts))
-        embeddings = embedding_func(embedding_texts)
-
-        for batch in create_batches(
-            api=CHROMA_CLIENT,
-            ids=[str(uuid.uuid4()) for _ in texts],
-            metadatas=metadatas,
-            embeddings=embeddings,
-            documents=texts,
-        ):
-            collection.add(*batch)
-
-        return True
+            return True
     except Exception as e:
         log.exception(e)
-        if e.__class__.__name__ == "UniqueConstraintError":
-            return True
-
         return False
 
 
+class TikaLoader:
+    def __init__(self, file_path, mime_type=None):
+        self.file_path = file_path
+        self.mime_type = mime_type
+
+    def load(self) -> list[Document]:
+        with open(self.file_path, "rb") as f:
+            data = f.read()
+
+        if self.mime_type is not None:
+            headers = {"Content-Type": self.mime_type}
+        else:
+            headers = {}
+
+        endpoint = app.state.config.TIKA_SERVER_URL
+        if not endpoint.endswith("/"):
+            endpoint += "/"
+        endpoint += "tika/text"
+
+        r = requests.put(endpoint, data=data, headers=headers)
+
+        if r.ok:
+            raw_metadata = r.json()
+            text = raw_metadata.get("X-TIKA:content", "<No text content found>")
+
+            if "Content-Type" in raw_metadata:
+                headers["Content-Type"] = raw_metadata["Content-Type"]
+
+            log.info("Tika extracted text: %s", text)
+
+            return [Document(page_content=text, metadata=headers)]
+        else:
+            raise Exception(f"Error calling Tika: {r.reason}")
+
+
 def get_loader(filename: str, file_content_type: str, file_path: str):
     file_ext = filename.split(".")[-1].lower()
     known_type = True
@@ -1033,49 +1214,67 @@ def get_loader(filename: str, file_content_type: str, file_path: str):
         "vue",
         "svelte",
         "msg",
+        "ex",
+        "exs",
+        "erl",
+        "tsx",
+        "jsx",
+        "hs",
+        "lhs",
     ]
 
-    if file_ext == "pdf":
-        loader = PyPDFLoader(
-            file_path, extract_images=app.state.config.PDF_EXTRACT_IMAGES
-        )
-    elif file_ext == "csv":
-        loader = CSVLoader(file_path)
-    elif file_ext == "rst":
-        loader = UnstructuredRSTLoader(file_path, mode="elements")
-    elif file_ext == "xml":
-        loader = UnstructuredXMLLoader(file_path)
-    elif file_ext in ["htm", "html"]:
-        loader = BSHTMLLoader(file_path, open_encoding="unicode_escape")
-    elif file_ext == "md":
-        loader = UnstructuredMarkdownLoader(file_path)
-    elif file_content_type == "application/epub+zip":
-        loader = UnstructuredEPubLoader(file_path)
-    elif (
-        file_content_type
-        == "application/vnd.openxmlformats-officedocument.wordprocessingml.document"
-        or file_ext in ["doc", "docx"]
+    if (
+        app.state.config.CONTENT_EXTRACTION_ENGINE == "tika"
+        and app.state.config.TIKA_SERVER_URL
     ):
-        loader = Docx2txtLoader(file_path)
-    elif file_content_type in [
-        "application/vnd.ms-excel",
-        "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
-    ] or file_ext in ["xls", "xlsx"]:
-        loader = UnstructuredExcelLoader(file_path)
-    elif file_content_type in [
-        "application/vnd.ms-powerpoint",
-        "application/vnd.openxmlformats-officedocument.presentationml.presentation",
-    ] or file_ext in ["ppt", "pptx"]:
-        loader = UnstructuredPowerPointLoader(file_path)
-    elif file_ext == "msg":
-        loader = OutlookMessageLoader(file_path)
-    elif file_ext in known_source_ext or (
-        file_content_type and file_content_type.find("text/") >= 0
-    ):
-        loader = TextLoader(file_path, autodetect_encoding=True)
+        if file_ext in known_source_ext or (
+            file_content_type and file_content_type.find("text/") >= 0
+        ):
+            loader = TextLoader(file_path, autodetect_encoding=True)
+        else:
+            loader = TikaLoader(file_path, file_content_type)
     else:
-        loader = TextLoader(file_path, autodetect_encoding=True)
-        known_type = False
+        if file_ext == "pdf":
+            loader = PyPDFLoader(
+                file_path, extract_images=app.state.config.PDF_EXTRACT_IMAGES
+            )
+        elif file_ext == "csv":
+            loader = CSVLoader(file_path)
+        elif file_ext == "rst":
+            loader = UnstructuredRSTLoader(file_path, mode="elements")
+        elif file_ext == "xml":
+            loader = UnstructuredXMLLoader(file_path)
+        elif file_ext in ["htm", "html"]:
+            loader = BSHTMLLoader(file_path, open_encoding="unicode_escape")
+        elif file_ext == "md":
+            loader = UnstructuredMarkdownLoader(file_path)
+        elif file_content_type == "application/epub+zip":
+            loader = UnstructuredEPubLoader(file_path)
+        elif (
+            file_content_type
+            == "application/vnd.openxmlformats-officedocument.wordprocessingml.document"
+            or file_ext == "docx"
+        ):
+            loader = Docx2txtLoader(file_path)
+        elif file_content_type in [
+            "application/vnd.ms-excel",
+            "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+        ] or file_ext in ["xls", "xlsx"]:
+            loader = UnstructuredExcelLoader(file_path)
+        elif file_content_type in [
+            "application/vnd.ms-powerpoint",
+            "application/vnd.openxmlformats-officedocument.presentationml.presentation",
+        ] or file_ext in ["ppt", "pptx"]:
+            loader = UnstructuredPowerPointLoader(file_path)
+        elif file_ext == "msg":
+            loader = OutlookMessageLoader(file_path)
+        elif file_ext in known_source_ext or (
+            file_content_type and file_content_type.find("text/") >= 0
+        ):
+            loader = TextLoader(file_path, autodetect_encoding=True)
+        else:
+            loader = TextLoader(file_path, autodetect_encoding=True)
+            known_type = False
 
     return loader, known_type
 
@@ -1101,7 +1300,7 @@ def store_doc(
             f.close()
 
         f = open(file_path, "rb")
-        if collection_name == None:
+        if collection_name is None:
             collection_name = calculate_sha256(f)[:63]
         f.close()
 
@@ -1154,7 +1353,7 @@ def process_doc(
         f = open(file_path, "rb")
 
         collection_name = form_data.collection_name
-        if collection_name == None:
+        if collection_name is None:
             collection_name = calculate_sha256(f)[:63]
         f.close()
 
@@ -1164,13 +1363,21 @@ def process_doc(
         data = loader.load()
 
         try:
-            result = store_data_in_vector_db(data, collection_name)
+            result = store_data_in_vector_db(
+                data,
+                collection_name,
+                {
+                    "file_id": form_data.file_id,
+                    "name": file.meta.get("name", file.filename),
+                },
+            )
 
             if result:
                 return {
                     "status": True,
                     "collection_name": collection_name,
                     "known_type": known_type,
+                    "filename": file.meta.get("name", file.filename),
                 }
         except Exception as e:
             raise HTTPException(
@@ -1202,9 +1409,8 @@ def store_text(
     form_data: TextRAGForm,
     user=Depends(get_verified_user),
 ):
-
     collection_name = form_data.collection_name
-    if collection_name == None:
+    if collection_name is None:
         collection_name = calculate_sha256_string(form_data.content)
 
     result = store_text_in_vector_db(
@@ -1247,7 +1453,7 @@ def scan_docs_dir(user=Depends(get_admin_user)):
                         sanitized_filename = sanitize_filename(filename)
                         doc = Documents.get_doc_by_name(sanitized_filename)
 
-                        if doc == None:
+                        if doc is None:
                             doc = Documents.insert_new_doc(
                                 user.id,
                                 DocumentForm(
@@ -1283,12 +1489,12 @@ def scan_docs_dir(user=Depends(get_admin_user)):
     return True
 
 
-@app.get("/reset/db")
+@app.post("/reset/db")
 def reset_vector_db(user=Depends(get_admin_user)):
-    CHROMA_CLIENT.reset()
+    VECTOR_DB_CLIENT.reset()
 
 
-@app.get("/reset/uploads")
+@app.post("/reset/uploads")
 def reset_upload_dir(user=Depends(get_admin_user)) -> bool:
     folder = f"{UPLOAD_DIR}"
     try:
@@ -1312,7 +1518,7 @@ def reset_upload_dir(user=Depends(get_admin_user)) -> bool:
     return True
 
 
-@app.get("/reset")
+@app.post("/reset")
 def reset(user=Depends(get_admin_user)) -> bool:
     folder = f"{UPLOAD_DIR}"
     for filename in os.listdir(folder):
@@ -1326,7 +1532,7 @@ def reset(user=Depends(get_admin_user)) -> bool:
             log.error("Failed to delete %s. Reason: %s" % (file_path, e))
 
     try:
-        CHROMA_CLIENT.reset()
+        VECTOR_DB_CLIENT.reset()
     except Exception as e:
         log.exception(e)
 

backend/open_webui/apps/rag/search/brave.py

@@ -1,16 +1,16 @@
 import logging
-from typing import List, Optional
-import requests
+from typing import Optional
 
-from apps.rag.search.main import SearchResult, get_filtered_results
-from config import SRC_LOG_LEVELS
+import requests
+from open_webui.apps.rag.search.main import SearchResult, get_filtered_results
+from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
 
 
 def search_brave(
-    api_key: str, query: str, count: int, filter_list: Optional[List[str]] = None
+    api_key: str, query: str, count: int, filter_list: Optional[list[str]] = None
 ) -> list[SearchResult]:
     """Search using Brave's Search API and return the results as a list of SearchResult objects.
 

backend/open_webui/apps/rag/search/duckduckgo.py

@@ -1,15 +1,16 @@
 import logging
-from typing import List, Optional
-from apps.rag.search.main import SearchResult, get_filtered_results
+from typing import Optional
+
+from open_webui.apps.rag.search.main import SearchResult, get_filtered_results
 from duckduckgo_search import DDGS
-from config import SRC_LOG_LEVELS
+from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
 
 
 def search_duckduckgo(
-    query: str, count: int, filter_list: Optional[List[str]] = None
+    query: str, count: int, filter_list: Optional[list[str]] = None
 ) -> list[SearchResult]:
     """
     Search using DuckDuckGo's Search API and return the results as a list of SearchResult objects.
@@ -18,7 +19,7 @@ def search_duckduckgo(
         count (int): The number of results to return
 
     Returns:
-        List[SearchResult]: A list of search results
+        list[SearchResult]: A list of search results
     """
     # Use the DDGS context manager to create a DDGS object
     with DDGS() as ddgs:

backend/open_webui/apps/rag/search/google_pse.py

@@ -1,10 +1,9 @@
-import json
 import logging
-from typing import List, Optional
-import requests
+from typing import Optional
 
-from apps.rag.search.main import SearchResult, get_filtered_results
-from config import SRC_LOG_LEVELS
+import requests
+from open_webui.apps.rag.search.main import SearchResult, get_filtered_results
+from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
@@ -15,7 +14,7 @@ def search_google_pse(
     search_engine_id: str,
     query: str,
     count: int,
-    filter_list: Optional[List[str]] = None,
+    filter_list: Optional[list[str]] = None,
 ) -> list[SearchResult]:
     """Search using Google's Programmable Search Engine API and return the results as a list of SearchResult objects.
 

backend/open_webui/apps/rag/search/jina_search.py

@@ -1,9 +1,9 @@
 import logging
-import requests
-from yarl import URL
 
-from apps.rag.search.main import SearchResult
-from config import SRC_LOG_LEVELS
+import requests
+from open_webui.apps.rag.search.main import SearchResult
+from open_webui.env import SRC_LOG_LEVELS
+from yarl import URL
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
@@ -17,7 +17,7 @@ def search_jina(query: str, count: int) -> list[SearchResult]:
         count (int): The number of results to return
 
     Returns:
-        List[SearchResult]: A list of search results
+        list[SearchResult]: A list of search results
     """
     jina_search_endpoint = "https://s.jina.ai/"
     headers = {

backend/open_webui/apps/rag/search/main.py

@@ -1,5 +1,6 @@
 from typing import Optional
 from urllib.parse import urlparse
+
 from pydantic import BaseModel
 
 
@@ -8,7 +9,8 @@ def get_filtered_results(results, filter_list):
         return results
     filtered_results = []
     for result in results:
-        domain = urlparse(result["url"]).netloc
+        url = result.get("url") or result.get("link", "")
+        domain = urlparse(url).netloc
         if any(domain.endswith(filtered_domain) for filtered_domain in filter_list):
             filtered_results.append(result)
     return filtered_results

backend/open_webui/apps/rag/search/searchapi.py

@@ -0,0 +1,48 @@
+import logging
+from typing import Optional
+from urllib.parse import urlencode
+
+import requests
+from open_webui.apps.rag.search.main import SearchResult, get_filtered_results
+from open_webui.env import SRC_LOG_LEVELS
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["RAG"])
+
+
+def search_searchapi(
+    api_key: str,
+    engine: str,
+    query: str,
+    count: int,
+    filter_list: Optional[list[str]] = None,
+) -> list[SearchResult]:
+    """Search using searchapi.io's API and return the results as a list of SearchResult objects.
+
+    Args:
+      api_key (str): A searchapi.io API key
+      query (str): The query to search for
+    """
+    url = "https://www.searchapi.io/api/v1/search"
+
+    engine = engine or "google"
+
+    payload = {"engine": engine, "q": query, "api_key": api_key}
+
+    url = f"{url}?{urlencode(payload)}"
+    response = requests.request("GET", url)
+
+    json_response = response.json()
+    log.info(f"results from searchapi search: {json_response}")
+
+    results = sorted(
+        json_response.get("organic_results", []), key=lambda x: x.get("position", 0)
+    )
+    if filter_list:
+        results = get_filtered_results(results, filter_list)
+    return [
+        SearchResult(
+            link=result["link"], title=result["title"], snippet=result["snippet"]
+        )
+        for result in results[:count]
+    ]

backend/open_webui/apps/rag/search/searxng.py

@@ -1,10 +1,9 @@
 import logging
+from typing import Optional
+
 import requests
-
-from typing import List, Optional
-
-from apps.rag.search.main import SearchResult, get_filtered_results
-from config import SRC_LOG_LEVELS
+from open_webui.apps.rag.search.main import SearchResult, get_filtered_results
+from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
@@ -14,9 +13,9 @@ def search_searxng(
     query_url: str,
     query: str,
     count: int,
-    filter_list: Optional[List[str]] = None,
+    filter_list: Optional[list[str]] = None,
     **kwargs,
-) -> List[SearchResult]:
+) -> list[SearchResult]:
     """
     Search a SearXNG instance for a given query and return the results as a list of SearchResult objects.
 
@@ -31,10 +30,10 @@ def search_searxng(
         language (str): Language filter for the search results; e.g., "en-US". Defaults to an empty string.
         safesearch (int): Safe search filter for safer web results; 0 = off, 1 = moderate, 2 = strict. Defaults to 1 (moderate).
         time_range (str): Time range for filtering results by date; e.g., "2023-04-05..today" or "all-time". Defaults to ''.
-        categories: (Optional[List[str]]): Specific categories within which the search should be performed, defaulting to an empty string if not provided.
+        categories: (Optional[list[str]]): Specific categories within which the search should be performed, defaulting to an empty string if not provided.
 
     Returns:
-        List[SearchResult]: A list of SearchResults sorted by relevance score in descending order.
+        list[SearchResult]: A list of SearchResults sorted by relevance score in descending order.
 
     Raise:
         requests.exceptions.RequestException: If a request error occurs during the search process.

backend/open_webui/apps/rag/search/serper.py

@@ -1,17 +1,17 @@
 import json
 import logging
-from typing import List, Optional
-import requests
+from typing import Optional
 
-from apps.rag.search.main import SearchResult, get_filtered_results
-from config import SRC_LOG_LEVELS
+import requests
+from open_webui.apps.rag.search.main import SearchResult, get_filtered_results
+from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
 
 
 def search_serper(
-    api_key: str, query: str, count: int, filter_list: Optional[List[str]] = None
+    api_key: str, query: str, count: int, filter_list: Optional[list[str]] = None
 ) -> list[SearchResult]:
     """Search using serper.dev's API and return the results as a list of SearchResult objects.
 

backend/open_webui/apps/rag/search/serply.py

@@ -1,11 +1,10 @@
-import json
 import logging
-from typing import List, Optional
-import requests
+from typing import Optional
 from urllib.parse import urlencode
 
-from apps.rag.search.main import SearchResult, get_filtered_results
-from config import SRC_LOG_LEVELS
+import requests
+from open_webui.apps.rag.search.main import SearchResult, get_filtered_results
+from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
@@ -19,7 +18,7 @@ def search_serply(
     limit: int = 10,
     device_type: str = "desktop",
     proxy_location: str = "US",
-    filter_list: Optional[List[str]] = None,
+    filter_list: Optional[list[str]] = None,
 ) -> list[SearchResult]:
     """Search using serper.dev's API and return the results as a list of SearchResult objects.
 

backend/open_webui/apps/rag/search/serpstack.py

@@ -1,10 +1,9 @@
-import json
 import logging
-from typing import List, Optional
-import requests
+from typing import Optional
 
-from apps.rag.search.main import SearchResult, get_filtered_results
-from config import SRC_LOG_LEVELS
+import requests
+from open_webui.apps.rag.search.main import SearchResult, get_filtered_results
+from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
@@ -14,7 +13,7 @@ def search_serpstack(
     api_key: str,
     query: str,
     count: int,
-    filter_list: Optional[List[str]] = None,
+    filter_list: Optional[list[str]] = None,
     https_enabled: bool = True,
 ) -> list[SearchResult]:
     """Search using serpstack.com's and return the results as a list of SearchResult objects.

backend/open_webui/apps/rag/search/tavily.py

@@ -1,9 +1,8 @@
 import logging
 
 import requests
-
-from apps.rag.search.main import SearchResult
-from config import SRC_LOG_LEVELS
+from open_webui.apps.rag.search.main import SearchResult
+from open_webui.env import SRC_LOG_LEVELS
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
@@ -17,7 +16,7 @@ def search_tavily(api_key: str, query: str, count: int) -> list[SearchResult]:
         query (str): The query to search for
 
     Returns:
-        List[SearchResult]: A list of search results
+        list[SearchResult]: A list of search results
     """
     url = "https://api.tavily.com/search"
     data = {"query": query, "api_key": api_key}

backend/open_webui/apps/rag/utils.py

@@ -1,32 +1,68 @@
-import os
 import logging
+import os
+import uuid
+from typing import Optional, Union
+
 import requests
 
-from typing import List, Union
-
-from apps.ollama.main import (
-    generate_ollama_embeddings,
-    GenerateEmbeddingsForm,
-)
-
 from huggingface_hub import snapshot_download
-
-from langchain_core.documents import Document
+from langchain.retrievers import ContextualCompressionRetriever, EnsembleRetriever
 from langchain_community.retrievers import BM25Retriever
-from langchain.retrievers import (
-    ContextualCompressionRetriever,
-    EnsembleRetriever,
+from langchain_core.documents import Document
+
+
+from open_webui.apps.ollama.main import (
+    GenerateEmbeddingsForm,
+    generate_ollama_embeddings,
 )
+from open_webui.apps.rag.vector.connector import VECTOR_DB_CLIENT
+from open_webui.utils.misc import get_last_user_message
 
-from typing import Optional
+from open_webui.env import SRC_LOG_LEVELS
 
-from utils.misc import get_last_user_message, add_or_update_system_message
-from config import SRC_LOG_LEVELS, CHROMA_CLIENT
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["RAG"])
 
 
+from typing import Any
+
+from langchain_core.callbacks import CallbackManagerForRetrieverRun
+from langchain_core.retrievers import BaseRetriever
+
+
+class VectorSearchRetriever(BaseRetriever):
+    collection_name: Any
+    embedding_function: Any
+    top_k: int
+
+    def _get_relevant_documents(
+        self,
+        query: str,
+        *,
+        run_manager: CallbackManagerForRetrieverRun,
+    ) -> list[Document]:
+        result = VECTOR_DB_CLIENT.search(
+            collection_name=self.collection_name,
+            vectors=[self.embedding_function(query)],
+            limit=self.top_k,
+        )
+
+        ids = result.ids[0]
+        metadatas = result.metadatas[0]
+        documents = result.documents[0]
+
+        results = []
+        for idx in range(len(ids)):
+            results.append(
+                Document(
+                    metadata=metadatas[idx],
+                    page_content=documents[idx],
+                )
+            )
+        return results
+
+
 def query_doc(
     collection_name: str,
     query: str,
@@ -34,17 +70,18 @@ def query_doc(
     k: int,
 ):
     try:
-        collection = CHROMA_CLIENT.get_collection(name=collection_name)
-        query_embeddings = embedding_function(query)
-
-        result = collection.query(
-            query_embeddings=[query_embeddings],
-            n_results=k,
+        result = VECTOR_DB_CLIENT.search(
+            collection_name=collection_name,
+            vectors=[embedding_function(query)],
+            limit=k,
         )
 
+        print("result", result)
+
         log.info(f"query_doc:result {result}")
         return result
     except Exception as e:
+        print(e)
         raise e
 
 
@@ -55,27 +92,25 @@ def query_doc_with_hybrid_search(
     k: int,
     reranking_function,
     r: float,
-):
+) -> dict:
     try:
-        collection = CHROMA_CLIENT.get_collection(name=collection_name)
-        documents = collection.get()  # get all documents
+        result = VECTOR_DB_CLIENT.get(collection_name=collection_name)
 
         bm25_retriever = BM25Retriever.from_texts(
-            texts=documents.get("documents"),
-            metadatas=documents.get("metadatas"),
+            texts=result.documents[0],
+            metadatas=result.metadatas[0],
         )
         bm25_retriever.k = k
 
-        chroma_retriever = ChromaRetriever(
-            collection=collection,
+        vector_search_retriever = VectorSearchRetriever(
+            collection_name=collection_name,
             embedding_function=embedding_function,
-            top_n=k,
+            top_k=k,
         )
 
         ensemble_retriever = EnsembleRetriever(
-            retrievers=[bm25_retriever, chroma_retriever], weights=[0.5, 0.5]
+            retrievers=[bm25_retriever, vector_search_retriever], weights=[0.5, 0.5]
         )
-
         compressor = RerankCompressor(
             embedding_function=embedding_function,
             top_n=k,
@@ -100,7 +135,9 @@ def query_doc_with_hybrid_search(
         raise e
 
 
-def merge_and_sort_query_results(query_results, k, reverse=False):
+def merge_and_sort_query_results(
+    query_results: list[dict], k: int, reverse: bool = False
+) -> list[dict]:
     # Initialize lists to store combined data
     combined_distances = []
     combined_documents = []
@@ -142,35 +179,40 @@ def merge_and_sort_query_results(query_results, k, reverse=False):
 
 
 def query_collection(
-    collection_names: List[str],
+    collection_names: list[str],
     query: str,
     embedding_function,
     k: int,
-):
+) -> dict:
     results = []
     for collection_name in collection_names:
-        try:
-            result = query_doc(
-                collection_name=collection_name,
-                query=query,
-                k=k,
-                embedding_function=embedding_function,
-            )
-            results.append(result)
-        except:
+        if collection_name:
+            try:
+                result = query_doc(
+                    collection_name=collection_name,
+                    query=query,
+                    k=k,
+                    embedding_function=embedding_function,
+                )
+                results.append(result.model_dump())
+            except Exception as e:
+                log.exception(f"Error when querying the collection: {e}")
+        else:
             pass
+
     return merge_and_sort_query_results(results, k=k)
 
 
 def query_collection_with_hybrid_search(
-    collection_names: List[str],
+    collection_names: list[str],
     query: str,
     embedding_function,
     k: int,
     reranking_function,
     r: float,
-):
+) -> dict:
     results = []
+    error = False
     for collection_name in collection_names:
         try:
             result = query_doc_with_hybrid_search(
@@ -182,14 +224,39 @@ def query_collection_with_hybrid_search(
                 r=r,
             )
             results.append(result)
-        except:
-            pass
+        except Exception as e:
+            log.exception(
+                "Error when querying the collection with " f"hybrid_search: {e}"
+            )
+            error = True
+
+    if error:
+        raise Exception(
+            "Hybrid search failed for all collections. Using Non hybrid search as fallback."
+        )
+
     return merge_and_sort_query_results(results, k=k, reverse=True)
 
 
 def rag_template(template: str, context: str, query: str):
-    template = template.replace("[context]", context)
-    template = template.replace("[query]", query)
+    count = template.count("[context]")
+    assert "[context]" in template, "RAG template does not contain '[context]'"
+
+    if "<context>" in context and "</context>" in context:
+        log.debug(
+            "WARNING: Potential prompt injection attack: the RAG "
+            "context contains '<context>' and '</context>'. This might be "
+            "nothing, or the user might be trying to hack something."
+        )
+
+    if "[query]" in context:
+        query_placeholder = f"[query-{str(uuid.uuid4())}]"
+        template = template.replace("[query]", query_placeholder)
+        template = template.replace("[context]", context)
+        template = template.replace(query_placeholder, query)
+    else:
+        template = template.replace("[context]", context)
+        template = template.replace("[query]", query)
     return template
 
 
@@ -257,7 +324,7 @@ def get_rag_context(
         collection_names = (
             file["collection_names"]
             if file["type"] == "collection"
-            else [file["collection_name"]]
+            else [file["collection_name"]] if file["collection_name"] else []
         )
 
         collection_names = set(collection_names).difference(extracted_collections)
@@ -266,19 +333,27 @@ def get_rag_context(
             continue
 
         try:
+            context = None
             if file["type"] == "text":
                 context = file["content"]
             else:
                 if hybrid_search:
-                    context = query_collection_with_hybrid_search(
-                        collection_names=collection_names,
-                        query=query,
-                        embedding_function=embedding_function,
-                        k=k,
-                        reranking_function=reranking_function,
-                        r=r,
-                    )
-                else:
+                    try:
+                        context = query_collection_with_hybrid_search(
+                            collection_names=collection_names,
+                            query=query,
+                            embedding_function=embedding_function,
+                            k=k,
+                            reranking_function=reranking_function,
+                            r=r,
+                        )
+                    except Exception as e:
+                        log.debug(
+                            "Error when using hybrid search, using"
+                            " non hybrid search as fallback."
+                        )
+
+                if (not hybrid_search) or (context is None):
                     context = query_collection(
                         collection_names=collection_names,
                         query=query,
@@ -287,21 +362,22 @@ def get_rag_context(
                     )
         except Exception as e:
             log.exception(e)
-            context = None
 
         if context:
             relevant_contexts.append({**context, "source": file})
 
         extracted_collections.extend(collection_names)
 
-    context_string = ""
-
+    contexts = []
     citations = []
+
     for context in relevant_contexts:
         try:
             if "documents" in context:
-                context_string += "\n\n".join(
-                    [text for text in context["documents"][0] if text is not None]
+                contexts.append(
+                    "\n\n".join(
+                        [text for text in context["documents"][0] if text is not None]
+                    )
                 )
 
                 if "metadatas" in context:
@@ -315,9 +391,7 @@ def get_rag_context(
         except Exception as e:
             log.exception(e)
 
-    context_string = context_string.strip()
-
-    return context_string, citations
+    return contexts, citations
 
 
 def get_model_path(model: str, update_model: bool = False):
@@ -395,54 +469,11 @@ def generate_openai_batch_embeddings(
         return None
 
 
-from typing import Any
-
-from langchain_core.retrievers import BaseRetriever
-from langchain_core.callbacks import CallbackManagerForRetrieverRun
-
-
-class ChromaRetriever(BaseRetriever):
-    collection: Any
-    embedding_function: Any
-    top_n: int
-
-    def _get_relevant_documents(
-        self,
-        query: str,
-        *,
-        run_manager: CallbackManagerForRetrieverRun,
-    ) -> List[Document]:
-        query_embeddings = self.embedding_function(query)
-
-        results = self.collection.query(
-            query_embeddings=[query_embeddings],
-            n_results=self.top_n,
-        )
-
-        ids = results["ids"][0]
-        metadatas = results["metadatas"][0]
-        documents = results["documents"][0]
-
-        results = []
-        for idx in range(len(ids)):
-            results.append(
-                Document(
-                    metadata=metadatas[idx],
-                    page_content=documents[idx],
-                )
-            )
-        return results
-
-
 import operator
-
 from typing import Optional, Sequence
 
-from langchain_core.documents import BaseDocumentCompressor, Document
 from langchain_core.callbacks import Callbacks
-from langchain_core.pydantic_v1 import Extra
-
-from sentence_transformers import util
+from langchain_core.documents import BaseDocumentCompressor, Document
 
 
 class RerankCompressor(BaseDocumentCompressor):
@@ -452,7 +483,7 @@ class RerankCompressor(BaseDocumentCompressor):
     r_score: float
 
     class Config:
-        extra = Extra.forbid
+        extra = "forbid"
         arbitrary_types_allowed = True
 
     def compress_documents(
@@ -468,6 +499,8 @@ class RerankCompressor(BaseDocumentCompressor):
                 [(query, doc.page_content) for doc in documents]
             )
         else:
+            from sentence_transformers import util
+
             query_embedding = self.embedding_function(query)
             document_embedding = self.embedding_function(
                 [doc.page_content for doc in documents]

backend/open_webui/apps/rag/vector/connector.py

@@ -0,0 +1,10 @@
+from open_webui.apps.rag.vector.dbs.chroma import ChromaClient
+from open_webui.apps.rag.vector.dbs.milvus import MilvusClient
+
+
+from open_webui.config import VECTOR_DB
+
+if VECTOR_DB == "milvus":
+    VECTOR_DB_CLIENT = MilvusClient()
+else:
+    VECTOR_DB_CLIENT = ChromaClient()

backend/open_webui/apps/rag/vector/dbs/chroma.py

@@ -0,0 +1,122 @@
+import chromadb
+from chromadb import Settings
+from chromadb.utils.batch_utils import create_batches
+
+from typing import Optional
+
+from open_webui.apps.rag.vector.main import VectorItem, SearchResult, GetResult
+from open_webui.config import (
+    CHROMA_DATA_PATH,
+    CHROMA_HTTP_HOST,
+    CHROMA_HTTP_PORT,
+    CHROMA_HTTP_HEADERS,
+    CHROMA_HTTP_SSL,
+    CHROMA_TENANT,
+    CHROMA_DATABASE,
+)
+
+
+class ChromaClient:
+    def __init__(self):
+        if CHROMA_HTTP_HOST != "":
+            self.client = chromadb.HttpClient(
+                host=CHROMA_HTTP_HOST,
+                port=CHROMA_HTTP_PORT,
+                headers=CHROMA_HTTP_HEADERS,
+                ssl=CHROMA_HTTP_SSL,
+                tenant=CHROMA_TENANT,
+                database=CHROMA_DATABASE,
+                settings=Settings(allow_reset=True, anonymized_telemetry=False),
+            )
+        else:
+            self.client = chromadb.PersistentClient(
+                path=CHROMA_DATA_PATH,
+                settings=Settings(allow_reset=True, anonymized_telemetry=False),
+                tenant=CHROMA_TENANT,
+                database=CHROMA_DATABASE,
+            )
+
+    def has_collection(self, collection_name: str) -> bool:
+        # Check if the collection exists based on the collection name.
+        collections = self.client.list_collections()
+        return collection_name in [collection.name for collection in collections]
+
+    def delete_collection(self, collection_name: str):
+        # Delete the collection based on the collection name.
+        return self.client.delete_collection(name=collection_name)
+
+    def search(
+        self, collection_name: str, vectors: list[list[float | int]], limit: int
+    ) -> Optional[SearchResult]:
+        # Search for the nearest neighbor items based on the vectors and return 'limit' number of results.
+        collection = self.client.get_collection(name=collection_name)
+        if collection:
+            result = collection.query(
+                query_embeddings=vectors,
+                n_results=limit,
+            )
+
+            return SearchResult(
+                **{
+                    "ids": result["ids"],
+                    "distances": result["distances"],
+                    "documents": result["documents"],
+                    "metadatas": result["metadatas"],
+                }
+            )
+        return None
+
+    def get(self, collection_name: str) -> Optional[GetResult]:
+        # Get all the items in the collection.
+        collection = self.client.get_collection(name=collection_name)
+        if collection:
+            result = collection.get()
+            return GetResult(
+                **{
+                    "ids": [result["ids"]],
+                    "documents": [result["documents"]],
+                    "metadatas": [result["metadatas"]],
+                }
+            )
+        return None
+
+    def insert(self, collection_name: str, items: list[VectorItem]):
+        # Insert the items into the collection, if the collection does not exist, it will be created.
+        collection = self.client.get_or_create_collection(name=collection_name)
+
+        ids = [item["id"] for item in items]
+        documents = [item["text"] for item in items]
+        embeddings = [item["vector"] for item in items]
+        metadatas = [item["metadata"] for item in items]
+
+        for batch in create_batches(
+            api=self.client,
+            documents=documents,
+            embeddings=embeddings,
+            ids=ids,
+            metadatas=metadatas,
+        ):
+            collection.add(*batch)
+
+    def upsert(self, collection_name: str, items: list[VectorItem]):
+        # Update the items in the collection, if the items are not present, insert them. If the collection does not exist, it will be created.
+        collection = self.client.get_or_create_collection(name=collection_name)
+
+        ids = [item["id"] for item in items]
+        documents = [item["text"] for item in items]
+        embeddings = [item["vector"] for item in items]
+        metadatas = [item["metadata"] for item in items]
+
+        collection.upsert(
+            ids=ids, documents=documents, embeddings=embeddings, metadatas=metadatas
+        )
+
+    def delete(self, collection_name: str, ids: list[str]):
+        # Delete the items from the collection based on the ids.
+        collection = self.client.get_collection(name=collection_name)
+        if collection:
+            collection.delete(ids=ids)
+
+    def reset(self):
+        # Resets the database. This will delete all collections and item entries.
+        return self.client.reset()

backend/open_webui/apps/rag/vector/dbs/milvus.py

@@ -0,0 +1,205 @@
+from pymilvus import MilvusClient as Client
+from pymilvus import FieldSchema, DataType
+import json
+
+from typing import Optional
+
+from open_webui.apps.rag.vector.main import VectorItem, SearchResult, GetResult
+from open_webui.config import (
+    MILVUS_URI,
+)
+
+
+class MilvusClient:
+    def __init__(self):
+        self.collection_prefix = "open_webui"
+        self.client = Client(uri=MILVUS_URI)
+
+    def _result_to_get_result(self, result) -> GetResult:
+        print(result)
+
+        ids = []
+        documents = []
+        metadatas = []
+
+        for match in result:
+            _ids = []
+            _documents = []
+            _metadatas = []
+
+            for item in match:
+                _ids.append(item.get("id"))
+                _documents.append(item.get("data", {}).get("text"))
+                _metadatas.append(item.get("metadata"))
+
+            ids.append(_ids)
+            documents.append(_documents)
+            metadatas.append(_metadatas)
+
+        return GetResult(
+            **{
+                "ids": ids,
+                "documents": documents,
+                "metadatas": metadatas,
+            }
+        )
+
+    def _result_to_search_result(self, result) -> SearchResult:
+        print(result)
+
+        ids = []
+        distances = []
+        documents = []
+        metadatas = []
+
+        for match in result:
+            _ids = []
+            _distances = []
+            _documents = []
+            _metadatas = []
+
+            for item in match:
+                _ids.append(item.get("id"))
+                _distances.append(item.get("distance"))
+                _documents.append(item.get("entity", {}).get("data", {}).get("text"))
+                _metadatas.append(item.get("entity", {}).get("metadata"))
+
+            ids.append(_ids)
+            distances.append(_distances)
+            documents.append(_documents)
+            metadatas.append(_metadatas)
+
+        return SearchResult(
+            **{
+                "ids": ids,
+                "distances": distances,
+                "documents": documents,
+                "metadatas": metadatas,
+            }
+        )
+
+    def _create_collection(self, collection_name: str, dimension: int):
+        schema = self.client.create_schema(
+            auto_id=False,
+            enable_dynamic_field=True,
+        )
+        schema.add_field(
+            field_name="id",
+            datatype=DataType.VARCHAR,
+            is_primary=True,
+            max_length=65535,
+        )
+        schema.add_field(
+            field_name="vector",
+            datatype=DataType.FLOAT_VECTOR,
+            dim=dimension,
+            description="vector",
+        )
+        schema.add_field(field_name="data", datatype=DataType.JSON, description="data")
+        schema.add_field(
+            field_name="metadata", datatype=DataType.JSON, description="metadata"
+        )
+
+        index_params = self.client.prepare_index_params()
+        index_params.add_index(
+            field_name="vector", index_type="HNSW", metric_type="COSINE", params={}
+        )
+
+        self.client.create_collection(
+            collection_name=f"{self.collection_prefix}_{collection_name}",
+            schema=schema,
+            index_params=index_params,
+        )
+
+    def has_collection(self, collection_name: str) -> bool:
+        # Check if the collection exists based on the collection name.
+        return self.client.has_collection(
+            collection_name=f"{self.collection_prefix}_{collection_name}"
+        )
+
+    def delete_collection(self, collection_name: str):
+        # Delete the collection based on the collection name.
+        return self.client.drop_collection(
+            collection_name=f"{self.collection_prefix}_{collection_name}"
+        )
+
+    def search(
+        self, collection_name: str, vectors: list[list[float | int]], limit: int
+    ) -> Optional[SearchResult]:
+        # Search for the nearest neighbor items based on the vectors and return 'limit' number of results.
+        result = self.client.search(
+            collection_name=f"{self.collection_prefix}_{collection_name}",
+            data=vectors,
+            limit=limit,
+            output_fields=["data", "metadata"],
+        )
+
+        return self._result_to_search_result(result)
+
+    def get(self, collection_name: str) -> Optional[GetResult]:
+        # Get all the items in the collection.
+        result = self.client.query(
+            collection_name=f"{self.collection_prefix}_{collection_name}",
+            filter='id != ""',
+        )
+        return self._result_to_get_result([result])
+
+    def insert(self, collection_name: str, items: list[VectorItem]):
+        # Insert the items into the collection, if the collection does not exist, it will be created.
+        if not self.client.has_collection(
+            collection_name=f"{self.collection_prefix}_{collection_name}"
+        ):
+            self._create_collection(
+                collection_name=collection_name, dimension=len(items[0]["vector"])
+            )
+
+        return self.client.insert(
+            collection_name=f"{self.collection_prefix}_{collection_name}",
+            data=[
+                {
+                    "id": item["id"],
+                    "vector": item["vector"],
+                    "data": {"text": item["text"]},
+                    "metadata": item["metadata"],
+                }
+                for item in items
+            ],
+        )
+
+    def upsert(self, collection_name: str, items: list[VectorItem]):
+        # Update the items in the collection, if the items are not present, insert them. If the collection does not exist, it will be created.
+        if not self.client.has_collection(
+            collection_name=f"{self.collection_prefix}_{collection_name}"
+        ):
+            self._create_collection(
+                collection_name=collection_name, dimension=len(items[0]["vector"])
+            )
+
+        return self.client.upsert(
+            collection_name=f"{self.collection_prefix}_{collection_name}",
+            data=[
+                {
+                    "id": item["id"],
+                    "vector": item["vector"],
+                    "data": {"text": item["text"]},
+                    "metadata": item["metadata"],
+                }
+                for item in items
+            ],
+        )
+
+    def delete(self, collection_name: str, ids: list[str]):
+        # Delete the items from the collection based on the ids.
+
+        return self.client.delete(
+            collection_name=f"{self.collection_prefix}_{collection_name}",
+            ids=ids,
+        )
+
+    def reset(self):
+        # Resets the database. This will delete all collections and item entries.
+
+        collection_names = self.client.list_collections()
+        for collection_name in collection_names:
+            if collection_name.startswith(self.collection_prefix):
+                self.client.drop_collection(collection_name=collection_name)

backend/open_webui/apps/rag/vector/main.py

@@ -0,0 +1,19 @@
+from pydantic import BaseModel
+from typing import Optional, List, Any
+
+
+class VectorItem(BaseModel):
+    id: str
+    text: str
+    vector: List[float | int]
+    metadata: Any
+
+
+class GetResult(BaseModel):
+    ids: Optional[List[List[str]]]
+    documents: Optional[List[List[str]]]
+    metadatas: Optional[List[List[Any]]]
+
+
+class SearchResult(GetResult):
+    distances: Optional[List[List[float | int]]]

backend/open_webui/apps/socket/main.py

@@ -0,0 +1,219 @@
+import asyncio
+import socketio
+import logging
+import sys
+import time
+
+from open_webui.apps.webui.models.users import Users
+from open_webui.env import (
+    ENABLE_WEBSOCKET_SUPPORT,
+    WEBSOCKET_MANAGER,
+    WEBSOCKET_REDIS_URL,
+)
+from open_webui.utils.utils import decode_token
+from open_webui.apps.socket.utils import RedisDict
+
+from open_webui.env import (
+    GLOBAL_LOG_LEVEL,
+    SRC_LOG_LEVELS,
+)
+
+
+logging.basicConfig(stream=sys.stdout, level=GLOBAL_LOG_LEVEL)
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["SOCKET"])
+
+
+if WEBSOCKET_MANAGER == "redis":
+    mgr = socketio.AsyncRedisManager(WEBSOCKET_REDIS_URL)
+    sio = socketio.AsyncServer(
+        cors_allowed_origins=[],
+        async_mode="asgi",
+        transports=(
+            ["polling", "websocket"] if ENABLE_WEBSOCKET_SUPPORT else ["polling"]
+        ),
+        allow_upgrades=ENABLE_WEBSOCKET_SUPPORT,
+        always_connect=True,
+        client_manager=mgr,
+    )
+else:
+    sio = socketio.AsyncServer(
+        cors_allowed_origins=[],
+        async_mode="asgi",
+        transports=(
+            ["polling", "websocket"] if ENABLE_WEBSOCKET_SUPPORT else ["polling"]
+        ),
+        allow_upgrades=ENABLE_WEBSOCKET_SUPPORT,
+        always_connect=True,
+    )
+
+
+# Dictionary to maintain the user pool
+
+if WEBSOCKET_MANAGER == "redis":
+    SESSION_POOL = RedisDict("open-webui:session_pool", redis_url=WEBSOCKET_REDIS_URL)
+    USER_POOL = RedisDict("open-webui:user_pool", redis_url=WEBSOCKET_REDIS_URL)
+    USAGE_POOL = RedisDict("open-webui:usage_pool", redis_url=WEBSOCKET_REDIS_URL)
+else:
+    SESSION_POOL = {}
+    USER_POOL = {}
+    USAGE_POOL = {}
+
+
+# Timeout duration in seconds
+TIMEOUT_DURATION = 3
+
+
+async def periodic_usage_pool_cleanup():
+    while True:
+        now = int(time.time())
+        for model_id, connections in list(USAGE_POOL.items()):
+            # Creating a list of sids to remove if they have timed out
+            expired_sids = [
+                sid
+                for sid, details in connections.items()
+                if now - details["updated_at"] > TIMEOUT_DURATION
+            ]
+
+            for sid in expired_sids:
+                del connections[sid]
+
+            if not connections:
+                log.debug(f"Cleaning up model {model_id} from usage pool")
+                del USAGE_POOL[model_id]
+            else:
+                USAGE_POOL[model_id] = connections
+
+            # Emit updated usage information after cleaning
+            await sio.emit("usage", {"models": get_models_in_use()})
+
+        await asyncio.sleep(TIMEOUT_DURATION)
+
+
+app = socketio.ASGIApp(
+    sio,
+    socketio_path="/ws/socket.io",
+)
+
+
+def get_models_in_use():
+    # List models that are currently in use
+    models_in_use = list(USAGE_POOL.keys())
+    return models_in_use
+
+
+@sio.on("usage")
+async def usage(sid, data):
+    model_id = data["model"]
+    # Record the timestamp for the last update
+    current_time = int(time.time())
+
+    # Store the new usage data and task
+    USAGE_POOL[model_id] = {
+        **(USAGE_POOL[model_id] if model_id in USAGE_POOL else {}),
+        sid: {"updated_at": current_time},
+    }
+
+    # Broadcast the usage data to all clients
+    await sio.emit("usage", {"models": get_models_in_use()})
+
+
+@sio.event
+async def connect(sid, environ, auth):
+    user = None
+    if auth and "token" in auth:
+        data = decode_token(auth["token"])
+
+        if data is not None and "id" in data:
+            user = Users.get_user_by_id(data["id"])
+
+        if user:
+            SESSION_POOL[sid] = user.id
+            if user.id in USER_POOL:
+                USER_POOL[user.id].append(sid)
+            else:
+                USER_POOL[user.id] = [sid]
+
+            # print(f"user {user.name}({user.id}) connected with session ID {sid}")
+            await sio.emit("user-count", {"count": len(USER_POOL.items())})
+            await sio.emit("usage", {"models": get_models_in_use()})
+
+
+@sio.on("user-join")
+async def user_join(sid, data):
+    # print("user-join", sid, data)
+
+    auth = data["auth"] if "auth" in data else None
+    if not auth or "token" not in auth:
+        return
+
+    data = decode_token(auth["token"])
+    if data is None or "id" not in data:
+        return
+
+    user = Users.get_user_by_id(data["id"])
+    if not user:
+        return
+
+    SESSION_POOL[sid] = user.id
+    if user.id in USER_POOL:
+        USER_POOL[user.id].append(sid)
+    else:
+        USER_POOL[user.id] = [sid]
+
+    # print(f"user {user.name}({user.id}) connected with session ID {sid}")
+
+    await sio.emit("user-count", {"count": len(USER_POOL.items())})
+
+
+@sio.on("user-count")
+async def user_count(sid):
+    await sio.emit("user-count", {"count": len(USER_POOL.items())})
+
+
+@sio.event
+async def disconnect(sid):
+    if sid in SESSION_POOL:
+        user_id = SESSION_POOL[sid]
+        del SESSION_POOL[sid]
+
+        USER_POOL[user_id] = [_sid for _sid in USER_POOL[user_id] if _sid != sid]
+
+        if len(USER_POOL[user_id]) == 0:
+            del USER_POOL[user_id]
+
+        await sio.emit("user-count", {"count": len(USER_POOL)})
+    else:
+        pass
+        # print(f"Unknown session ID {sid} disconnected")
+
+
+def get_event_emitter(request_info):
+    async def __event_emitter__(event_data):
+        await sio.emit(
+            "chat-events",
+            {
+                "chat_id": request_info["chat_id"],
+                "message_id": request_info["message_id"],
+                "data": event_data,
+            },
+            to=request_info["session_id"],
+        )
+
+    return __event_emitter__
+
+
+def get_event_call(request_info):
+    async def __event_call__(event_data):
+        response = await sio.call(
+            "chat-events",
+            {
+                "chat_id": request_info["chat_id"],
+                "message_id": request_info["message_id"],
+                "data": event_data,
+            },
+            to=request_info["session_id"],
+        )
+        return response
+
+    return __event_call__

backend/open_webui/apps/socket/utils.py

@@ -0,0 +1,59 @@
+import json
+import redis
+
+
+class RedisDict:
+    def __init__(self, name, redis_url):
+        self.name = name
+        self.redis = redis.Redis.from_url(redis_url, decode_responses=True)
+
+    def __setitem__(self, key, value):
+        serialized_value = json.dumps(value)
+        self.redis.hset(self.name, key, serialized_value)
+
+    def __getitem__(self, key):
+        value = self.redis.hget(self.name, key)
+        if value is None:
+            raise KeyError(key)
+        return json.loads(value)
+
+    def __delitem__(self, key):
+        result = self.redis.hdel(self.name, key)
+        if result == 0:
+            raise KeyError(key)
+
+    def __contains__(self, key):
+        return self.redis.hexists(self.name, key)
+
+    def __len__(self):
+        return self.redis.hlen(self.name)
+
+    def keys(self):
+        return self.redis.hkeys(self.name)
+
+    def values(self):
+        return [json.loads(v) for v in self.redis.hvals(self.name)]
+
+    def items(self):
+        return [(k, json.loads(v)) for k, v in self.redis.hgetall(self.name).items()]
+
+    def get(self, key, default=None):
+        try:
+            return self[key]
+        except KeyError:
+            return default
+
+    def clear(self):
+        self.redis.delete(self.name)
+
+    def update(self, other=None, **kwargs):
+        if other is not None:
+            for k, v in other.items() if hasattr(other, "items") else other:
+                self[k] = v
+        for k, v in kwargs.items():
+            self[k] = v
+
+    def setdefault(self, key, default=None):
+        if key not in self:
+            self[key] = default
+        return self[key]

backend/open_webui/apps/webui/internal/db.py

@@ -0,0 +1,92 @@
+import json
+import logging
+from contextlib import contextmanager
+from typing import Any, Optional
+
+from open_webui.apps.webui.internal.wrappers import register_connection
+from open_webui.env import OPEN_WEBUI_DIR, DATABASE_URL, SRC_LOG_LEVELS
+from peewee_migrate import Router
+from sqlalchemy import Dialect, create_engine, types
+from sqlalchemy.ext.declarative import declarative_base
+from sqlalchemy.orm import scoped_session, sessionmaker
+from sqlalchemy.sql.type_api import _T
+from typing_extensions import Self
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["DB"])
+
+
+class JSONField(types.TypeDecorator):
+    impl = types.Text
+    cache_ok = True
+
+    def process_bind_param(self, value: Optional[_T], dialect: Dialect) -> Any:
+        return json.dumps(value)
+
+    def process_result_value(self, value: Optional[_T], dialect: Dialect) -> Any:
+        if value is not None:
+            return json.loads(value)
+
+    def copy(self, **kw: Any) -> Self:
+        return JSONField(self.impl.length)
+
+    def db_value(self, value):
+        return json.dumps(value)
+
+    def python_value(self, value):
+        if value is not None:
+            return json.loads(value)
+
+
+# Workaround to handle the peewee migration
+# This is required to ensure the peewee migration is handled before the alembic migration
+def handle_peewee_migration(DATABASE_URL):
+    # db = None
+    try:
+        # Replace the postgresql:// with postgres:// to handle the peewee migration
+        db = register_connection(DATABASE_URL.replace("postgresql://", "postgres://"))
+        migrate_dir = OPEN_WEBUI_DIR / "apps" / "webui" / "internal" / "migrations"
+        router = Router(db, logger=log, migrate_dir=migrate_dir)
+        router.run()
+        db.close()
+
+    except Exception as e:
+        log.error(f"Failed to initialize the database connection: {e}")
+        raise
+    finally:
+        # Properly closing the database connection
+        if db and not db.is_closed():
+            db.close()
+
+        # Assert if db connection has been closed
+        assert db.is_closed(), "Database connection is still open."
+
+
+handle_peewee_migration(DATABASE_URL)
+
+
+SQLALCHEMY_DATABASE_URL = DATABASE_URL
+if "sqlite" in SQLALCHEMY_DATABASE_URL:
+    engine = create_engine(
+        SQLALCHEMY_DATABASE_URL, connect_args={"check_same_thread": False}
+    )
+else:
+    engine = create_engine(SQLALCHEMY_DATABASE_URL, pool_pre_ping=True)
+
+
+SessionLocal = sessionmaker(
+    autocommit=False, autoflush=False, bind=engine, expire_on_commit=False
+)
+Base = declarative_base()
+Session = scoped_session(SessionLocal)
+
+
+def get_session():
+    db = SessionLocal()
+    try:
+        yield db
+    finally:
+        db.close()
+
+
+get_db = contextmanager(get_session)

backend/open_webui/apps/webui/internal/migrations/010_migrate_modelfiles_to_models.py

@@ -30,7 +30,7 @@ import peewee as pw
 from peewee_migrate import Migrator
 import json
 
-from utils.misc import parse_ollama_modelfile
+from open_webui.utils.misc import parse_ollama_modelfile
 
 with suppress(ImportError):
     import playhouse.postgres_ext as pw_pext

backend/open_webui/apps/webui/internal/migrations/017_add_user_oauth_sub.py

@@ -1,10 +1,7 @@
 """Peewee migrations -- 017_add_user_oauth_sub.py.
-
 Some examples (model - class or model name)::
-
     > Model = migrator.orm['table_name']            # Return model in current state by name
     > Model = migrator.ModelClass                   # Return model in current state by name
-
     > migrator.sql(sql)                             # Run custom SQL
     > migrator.run(func, *args, **kwargs)           # Run python function with the given args
     > migrator.create_model(Model)                  # Create a model (could be used as decorator)
@@ -21,7 +18,6 @@ Some examples (model - class or model name)::
     > migrator.drop_index(model, *col_names)
     > migrator.drop_not_null(model, *field_names)
     > migrator.drop_constraints(model, *constraints)
-
 """
 
 from contextlib import suppress

backend/open_webui/apps/webui/internal/wrappers.py

@@ -1,13 +1,13 @@
-from contextvars import ContextVar
-from peewee import *
-from peewee import PostgresqlDatabase, InterfaceError as PeeWeeInterfaceError
-
 import logging
+from contextvars import ContextVar
+
+from open_webui.env import SRC_LOG_LEVELS
+from peewee import *
+from peewee import InterfaceError as PeeWeeInterfaceError
+from peewee import PostgresqlDatabase
 from playhouse.db_url import connect, parse
 from playhouse.shortcuts import ReconnectMixin
 
-from config import SRC_LOG_LEVELS
-
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["DB"])
 
@@ -43,7 +43,7 @@ class ReconnectingPostgresqlDatabase(CustomReconnectMixin, PostgresqlDatabase):
 
 
 def register_connection(db_url):
-    db = connect(db_url)
+    db = connect(db_url, unquote_password=True)
     if isinstance(db, PostgresqlDatabase):
         # Enable autoconnect for SQLite databases, managed by Peewee
         db.autoconnect = True
@@ -51,16 +51,10 @@ def register_connection(db_url):
         log.info("Connected to PostgreSQL database")
 
         # Get the connection details
-        connection = parse(db_url)
+        connection = parse(db_url, unquote_password=True)
 
         # Use our custom database class that supports reconnection
-        db = ReconnectingPostgresqlDatabase(
-            connection["database"],
-            user=connection["user"],
-            password=connection["password"],
-            host=connection["host"],
-            port=connection["port"],
-        )
+        db = ReconnectingPostgresqlDatabase(**connection)
         db.connect(reuse_if_open=True)
     elif isinstance(db, SqliteDatabase):
         # Enable autoconnect for SQLite databases, managed by Peewee

backend/open_webui/apps/webui/main.py

@@ -0,0 +1,390 @@
+import inspect
+import json
+import logging
+from typing import AsyncGenerator, Generator, Iterator
+
+from open_webui.apps.socket.main import get_event_call, get_event_emitter
+from open_webui.apps.webui.models.functions import Functions
+from open_webui.apps.webui.models.models import Models
+from open_webui.apps.webui.routers import (
+    auths,
+    chats,
+    configs,
+    documents,
+    files,
+    functions,
+    memories,
+    models,
+    prompts,
+    tools,
+    users,
+    utils,
+)
+from open_webui.apps.webui.utils import load_function_module_by_id
+from open_webui.config import (
+    ADMIN_EMAIL,
+    CORS_ALLOW_ORIGIN,
+    DEFAULT_MODELS,
+    DEFAULT_PROMPT_SUGGESTIONS,
+    DEFAULT_USER_ROLE,
+    ENABLE_COMMUNITY_SHARING,
+    ENABLE_LOGIN_FORM,
+    ENABLE_MESSAGE_RATING,
+    ENABLE_SIGNUP,
+    JWT_EXPIRES_IN,
+    OAUTH_EMAIL_CLAIM,
+    OAUTH_PICTURE_CLAIM,
+    OAUTH_USERNAME_CLAIM,
+    SHOW_ADMIN_DETAILS,
+    USER_PERMISSIONS,
+    WEBHOOK_URL,
+    WEBUI_AUTH,
+    WEBUI_BANNERS,
+    AppConfig,
+)
+from open_webui.env import (
+    WEBUI_AUTH_TRUSTED_EMAIL_HEADER,
+    WEBUI_AUTH_TRUSTED_NAME_HEADER,
+)
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import StreamingResponse
+from pydantic import BaseModel
+from open_webui.utils.misc import (
+    openai_chat_chunk_message_template,
+    openai_chat_completion_message_template,
+)
+from open_webui.utils.payload import (
+    apply_model_params_to_body_openai,
+    apply_model_system_prompt_to_body,
+)
+
+
+from open_webui.utils.tools import get_tools
+
+app = FastAPI()
+
+log = logging.getLogger(__name__)
+
+app.state.config = AppConfig()
+
+app.state.config.ENABLE_SIGNUP = ENABLE_SIGNUP
+app.state.config.ENABLE_LOGIN_FORM = ENABLE_LOGIN_FORM
+app.state.config.JWT_EXPIRES_IN = JWT_EXPIRES_IN
+app.state.AUTH_TRUSTED_EMAIL_HEADER = WEBUI_AUTH_TRUSTED_EMAIL_HEADER
+app.state.AUTH_TRUSTED_NAME_HEADER = WEBUI_AUTH_TRUSTED_NAME_HEADER
+
+
+app.state.config.SHOW_ADMIN_DETAILS = SHOW_ADMIN_DETAILS
+app.state.config.ADMIN_EMAIL = ADMIN_EMAIL
+
+
+app.state.config.DEFAULT_MODELS = DEFAULT_MODELS
+app.state.config.DEFAULT_PROMPT_SUGGESTIONS = DEFAULT_PROMPT_SUGGESTIONS
+app.state.config.DEFAULT_USER_ROLE = DEFAULT_USER_ROLE
+app.state.config.USER_PERMISSIONS = USER_PERMISSIONS
+app.state.config.WEBHOOK_URL = WEBHOOK_URL
+app.state.config.BANNERS = WEBUI_BANNERS
+
+app.state.config.ENABLE_COMMUNITY_SHARING = ENABLE_COMMUNITY_SHARING
+app.state.config.ENABLE_MESSAGE_RATING = ENABLE_MESSAGE_RATING
+
+app.state.config.OAUTH_USERNAME_CLAIM = OAUTH_USERNAME_CLAIM
+app.state.config.OAUTH_PICTURE_CLAIM = OAUTH_PICTURE_CLAIM
+app.state.config.OAUTH_EMAIL_CLAIM = OAUTH_EMAIL_CLAIM
+
+app.state.MODELS = {}
+app.state.TOOLS = {}
+app.state.FUNCTIONS = {}
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=CORS_ALLOW_ORIGIN,
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+
+app.include_router(configs.router, prefix="/configs", tags=["configs"])
+app.include_router(auths.router, prefix="/auths", tags=["auths"])
+app.include_router(users.router, prefix="/users", tags=["users"])
+app.include_router(chats.router, prefix="/chats", tags=["chats"])
+
+app.include_router(documents.router, prefix="/documents", tags=["documents"])
+app.include_router(models.router, prefix="/models", tags=["models"])
+app.include_router(prompts.router, prefix="/prompts", tags=["prompts"])
+
+app.include_router(memories.router, prefix="/memories", tags=["memories"])
+app.include_router(files.router, prefix="/files", tags=["files"])
+app.include_router(tools.router, prefix="/tools", tags=["tools"])
+app.include_router(functions.router, prefix="/functions", tags=["functions"])
+
+app.include_router(utils.router, prefix="/utils", tags=["utils"])
+
+
+@app.get("/")
+async def get_status():
+    return {
+        "status": True,
+        "auth": WEBUI_AUTH,
+        "default_models": app.state.config.DEFAULT_MODELS,
+        "default_prompt_suggestions": app.state.config.DEFAULT_PROMPT_SUGGESTIONS,
+    }
+
+
+def get_function_module(pipe_id: str):
+    # Check if function is already loaded
+    if pipe_id not in app.state.FUNCTIONS:
+        function_module, _, _ = load_function_module_by_id(pipe_id)
+        app.state.FUNCTIONS[pipe_id] = function_module
+    else:
+        function_module = app.state.FUNCTIONS[pipe_id]
+
+    if hasattr(function_module, "valves") and hasattr(function_module, "Valves"):
+        valves = Functions.get_function_valves_by_id(pipe_id)
+        function_module.valves = function_module.Valves(**(valves if valves else {}))
+    return function_module
+
+
+async def get_pipe_models():
+    pipes = Functions.get_functions_by_type("pipe", active_only=True)
+    pipe_models = []
+
+    for pipe in pipes:
+        function_module = get_function_module(pipe.id)
+
+        # Check if function is a manifold
+        if hasattr(function_module, "pipes"):
+            sub_pipes = []
+
+            # Check if pipes is a function or a list
+
+            try:
+                if callable(function_module.pipes):
+                    sub_pipes = function_module.pipes()
+                else:
+                    sub_pipes = function_module.pipes
+            except Exception as e:
+                log.exception(e)
+                sub_pipes = []
+
+            print(sub_pipes)
+
+            for p in sub_pipes:
+                sub_pipe_id = f'{pipe.id}.{p["id"]}'
+                sub_pipe_name = p["name"]
+
+                if hasattr(function_module, "name"):
+                    sub_pipe_name = f"{function_module.name}{sub_pipe_name}"
+
+                pipe_flag = {"type": pipe.type}
+                pipe_models.append(
+                    {
+                        "id": sub_pipe_id,
+                        "name": sub_pipe_name,
+                        "object": "model",
+                        "created": pipe.created_at,
+                        "owned_by": "openai",
+                        "pipe": pipe_flag,
+                    }
+                )
+        else:
+            pipe_flag = {"type": "pipe"}
+
+            pipe_models.append(
+                {
+                    "id": pipe.id,
+                    "name": pipe.name,
+                    "object": "model",
+                    "created": pipe.created_at,
+                    "owned_by": "openai",
+                    "pipe": pipe_flag,
+                }
+            )
+
+    return pipe_models
+
+
+async def execute_pipe(pipe, params):
+    if inspect.iscoroutinefunction(pipe):
+        return await pipe(**params)
+    else:
+        return pipe(**params)
+
+
+async def get_message_content(res: str | Generator | AsyncGenerator) -> str:
+    if isinstance(res, str):
+        return res
+    if isinstance(res, Generator):
+        return "".join(map(str, res))
+    if isinstance(res, AsyncGenerator):
+        return "".join([str(stream) async for stream in res])
+
+
+def process_line(form_data: dict, line):
+    if isinstance(line, BaseModel):
+        line = line.model_dump_json()
+        line = f"data: {line}"
+    if isinstance(line, dict):
+        line = f"data: {json.dumps(line)}"
+
+    try:
+        line = line.decode("utf-8")
+    except Exception:
+        pass
+
+    if line.startswith("data:"):
+        return f"{line}\n\n"
+    else:
+        line = openai_chat_chunk_message_template(form_data["model"], line)
+        return f"data: {json.dumps(line)}\n\n"
+
+
+def get_pipe_id(form_data: dict) -> str:
+    pipe_id = form_data["model"]
+    if "." in pipe_id:
+        pipe_id, _ = pipe_id.split(".", 1)
+    print(pipe_id)
+    return pipe_id
+
+
+def get_function_params(function_module, form_data, user, extra_params=None):
+    if extra_params is None:
+        extra_params = {}
+
+    pipe_id = get_pipe_id(form_data)
+
+    # Get the signature of the function
+    sig = inspect.signature(function_module.pipe)
+    params = {"body": form_data} | {
+        k: v for k, v in extra_params.items() if k in sig.parameters
+    }
+
+    if "__user__" in params and hasattr(function_module, "UserValves"):
+        user_valves = Functions.get_user_valves_by_id_and_user_id(pipe_id, user.id)
+        try:
+            params["__user__"]["valves"] = function_module.UserValves(**user_valves)
+        except Exception as e:
+            log.exception(e)
+            params["__user__"]["valves"] = function_module.UserValves()
+
+    return params
+
+
+async def generate_function_chat_completion(form_data, user):
+    model_id = form_data.get("model")
+    model_info = Models.get_model_by_id(model_id)
+
+    metadata = form_data.pop("metadata", {})
+
+    files = metadata.get("files", [])
+    tool_ids = metadata.get("tool_ids", [])
+    # Check if tool_ids is None
+    if tool_ids is None:
+        tool_ids = []
+
+    __event_emitter__ = None
+    __event_call__ = None
+    __task__ = None
+
+    if metadata:
+        if all(k in metadata for k in ("session_id", "chat_id", "message_id")):
+            __event_emitter__ = get_event_emitter(metadata)
+            __event_call__ = get_event_call(metadata)
+        __task__ = metadata.get("task", None)
+
+    extra_params = {
+        "__event_emitter__": __event_emitter__,
+        "__event_call__": __event_call__,
+        "__task__": __task__,
+        "__files__": files,
+        "__user__": {
+            "id": user.id,
+            "email": user.email,
+            "name": user.name,
+            "role": user.role,
+        },
+    }
+    extra_params["__tools__"] = get_tools(
+        app,
+        tool_ids,
+        user,
+        {
+            **extra_params,
+            "__model__": app.state.MODELS[form_data["model"]],
+            "__messages__": form_data["messages"],
+            "__files__": files,
+        },
+    )
+
+    if model_info:
+        if model_info.base_model_id:
+            form_data["model"] = model_info.base_model_id
+
+        params = model_info.params.model_dump()
+        form_data = apply_model_params_to_body_openai(params, form_data)
+        form_data = apply_model_system_prompt_to_body(params, form_data, user)
+
+    pipe_id = get_pipe_id(form_data)
+    function_module = get_function_module(pipe_id)
+
+    pipe = function_module.pipe
+    params = get_function_params(function_module, form_data, user, extra_params)
+
+    if form_data["stream"]:
+
+        async def stream_content():
+            try:
+                res = await execute_pipe(pipe, params)
+
+                # Directly return if the response is a StreamingResponse
+                if isinstance(res, StreamingResponse):
+                    async for data in res.body_iterator:
+                        yield data
+                    return
+                if isinstance(res, dict):
+                    yield f"data: {json.dumps(res)}\n\n"
+                    return
+
+            except Exception as e:
+                print(f"Error: {e}")
+                yield f"data: {json.dumps({'error': {'detail':str(e)}})}\n\n"
+                return
+
+            if isinstance(res, str):
+                message = openai_chat_chunk_message_template(form_data["model"], res)
+                yield f"data: {json.dumps(message)}\n\n"
+
+            if isinstance(res, Iterator):
+                for line in res:
+                    yield process_line(form_data, line)
+
+            if isinstance(res, AsyncGenerator):
+                async for line in res:
+                    yield process_line(form_data, line)
+
+            if isinstance(res, str) or isinstance(res, Generator):
+                finish_message = openai_chat_chunk_message_template(
+                    form_data["model"], ""
+                )
+                finish_message["choices"][0]["finish_reason"] = "stop"
+                yield f"data: {json.dumps(finish_message)}\n\n"
+                yield "data: [DONE]"
+
+        return StreamingResponse(stream_content(), media_type="text/event-stream")
+    else:
+        try:
+            res = await execute_pipe(pipe, params)
+
+        except Exception as e:
+            print(f"Error: {e}")
+            return {"error": {"detail": str(e)}}
+
+        if isinstance(res, StreamingResponse) or isinstance(res, dict):
+            return res
+        if isinstance(res, BaseModel):
+            return res.model_dump()
+
+        message = await get_message_content(res)
+        return openai_chat_completion_message_template(form_data["model"], message)

backend/open_webui/apps/webui/models/auths.py

@@ -1,16 +1,13 @@
-from pydantic import BaseModel
-from typing import List, Union, Optional
-import time
-import uuid
 import logging
-from peewee import *
+import uuid
+from typing import Optional
 
-from apps.webui.models.users import UserModel, Users
-from utils.utils import verify_password
-
-from apps.webui.internal.db import DB
-
-from config import SRC_LOG_LEVELS
+from open_webui.apps.webui.internal.db import Base, get_db
+from open_webui.apps.webui.models.users import UserModel, Users
+from open_webui.env import SRC_LOG_LEVELS
+from pydantic import BaseModel
+from sqlalchemy import Boolean, Column, String, Text
+from open_webui.utils.utils import verify_password
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])
@@ -20,14 +17,13 @@ log.setLevel(SRC_LOG_LEVELS["MODELS"])
 ####################
 
 
-class Auth(Model):
-    id = CharField(unique=True)
-    email = CharField()
-    password = TextField()
-    active = BooleanField()
+class Auth(Base):
+    __tablename__ = "auth"
 
-    class Meta:
-        database = DB
+    id = Column(String, primary_key=True)
+    email = Column(String)
+    password = Column(Text)
+    active = Column(Boolean)
 
 
 class AuthModel(BaseModel):
@@ -94,10 +90,6 @@ class AddUserForm(SignupForm):
 
 
 class AuthsTable:
-    def __init__(self, db):
-        self.db = db
-        self.db.create_tables([Auth])
-
     def insert_new_auth(
         self,
         email: str,
@@ -107,37 +99,43 @@ class AuthsTable:
         role: str = "pending",
         oauth_sub: Optional[str] = None,
     ) -> Optional[UserModel]:
-        log.info("insert_new_auth")
+        with get_db() as db:
+            log.info("insert_new_auth")
 
-        id = str(uuid.uuid4())
+            id = str(uuid.uuid4())
 
-        auth = AuthModel(
-            **{"id": id, "email": email, "password": password, "active": True}
-        )
-        result = Auth.create(**auth.model_dump())
+            auth = AuthModel(
+                **{"id": id, "email": email, "password": password, "active": True}
+            )
+            result = Auth(**auth.model_dump())
+            db.add(result)
 
-        user = Users.insert_new_user(
-            id, name, email, profile_image_url, role, oauth_sub
-        )
+            user = Users.insert_new_user(
+                id, name, email, profile_image_url, role, oauth_sub
+            )
 
-        if result and user:
-            return user
-        else:
-            return None
+            db.commit()
+            db.refresh(result)
+
+            if result and user:
+                return user
+            else:
+                return None
 
     def authenticate_user(self, email: str, password: str) -> Optional[UserModel]:
         log.info(f"authenticate_user: {email}")
         try:
-            auth = Auth.get(Auth.email == email, Auth.active == True)
-            if auth:
-                if verify_password(password, auth.password):
-                    user = Users.get_user_by_id(auth.id)
-                    return user
+            with get_db() as db:
+                auth = db.query(Auth).filter_by(email=email, active=True).first()
+                if auth:
+                    if verify_password(password, auth.password):
+                        user = Users.get_user_by_id(auth.id)
+                        return user
+                    else:
+                        return None
                 else:
                     return None
-            else:
-                return None
-        except:
+        except Exception:
             return None
 
     def authenticate_user_by_api_key(self, api_key: str) -> Optional[UserModel]:
@@ -149,52 +147,55 @@ class AuthsTable:
         try:
             user = Users.get_user_by_api_key(api_key)
             return user if user else None
-        except:
+        except Exception:
             return False
 
     def authenticate_user_by_trusted_header(self, email: str) -> Optional[UserModel]:
         log.info(f"authenticate_user_by_trusted_header: {email}")
         try:
-            auth = Auth.get(Auth.email == email, Auth.active == True)
-            if auth:
-                user = Users.get_user_by_id(auth.id)
-                return user
-        except:
+            with get_db() as db:
+                auth = db.query(Auth).filter_by(email=email, active=True).first()
+                if auth:
+                    user = Users.get_user_by_id(auth.id)
+                    return user
+        except Exception:
             return None
 
     def update_user_password_by_id(self, id: str, new_password: str) -> bool:
         try:
-            query = Auth.update(password=new_password).where(Auth.id == id)
-            result = query.execute()
-
-            return True if result == 1 else False
-        except:
+            with get_db() as db:
+                result = (
+                    db.query(Auth).filter_by(id=id).update({"password": new_password})
+                )
+                db.commit()
+                return True if result == 1 else False
+        except Exception:
             return False
 
     def update_email_by_id(self, id: str, email: str) -> bool:
         try:
-            query = Auth.update(email=email).where(Auth.id == id)
-            result = query.execute()
-
-            return True if result == 1 else False
-        except:
+            with get_db() as db:
+                result = db.query(Auth).filter_by(id=id).update({"email": email})
+                db.commit()
+                return True if result == 1 else False
+        except Exception:
             return False
 
     def delete_auth_by_id(self, id: str) -> bool:
         try:
-            # Delete User
-            result = Users.delete_user_by_id(id)
+            with get_db() as db:
+                # Delete User
+                result = Users.delete_user_by_id(id)
 
-            if result:
-                # Delete Auth
-                query = Auth.delete().where(Auth.id == id)
-                query.execute()  # Remove the rows, return number of rows removed.
+                if result:
+                    db.query(Auth).filter_by(id=id).delete()
+                    db.commit()
 
-                return True
-            else:
-                return False
-        except:
+                    return True
+                else:
+                    return False
+        except Exception:
             return False
 
 
-Auths = AuthsTable(DB)
+Auths = AuthsTable()

backend/open_webui/apps/webui/models/chats.py

@@ -0,0 +1,386 @@
+import json
+import time
+import uuid
+from typing import Optional
+
+from open_webui.apps.webui.internal.db import Base, get_db
+from pydantic import BaseModel, ConfigDict
+from sqlalchemy import BigInteger, Boolean, Column, String, Text
+
+####################
+# Chat DB Schema
+####################
+
+
+class Chat(Base):
+    __tablename__ = "chat"
+
+    id = Column(String, primary_key=True)
+    user_id = Column(String)
+    title = Column(Text)
+    chat = Column(Text)  # Save Chat JSON as Text
+
+    created_at = Column(BigInteger)
+    updated_at = Column(BigInteger)
+
+    share_id = Column(Text, unique=True, nullable=True)
+    archived = Column(Boolean, default=False)
+
+
+class ChatModel(BaseModel):
+    model_config = ConfigDict(from_attributes=True)
+
+    id: str
+    user_id: str
+    title: str
+    chat: str
+
+    created_at: int  # timestamp in epoch
+    updated_at: int  # timestamp in epoch
+
+    share_id: Optional[str] = None
+    archived: bool = False
+
+
+####################
+# Forms
+####################
+
+
+class ChatForm(BaseModel):
+    chat: dict
+
+
+class ChatTitleForm(BaseModel):
+    title: str
+
+
+class ChatResponse(BaseModel):
+    id: str
+    user_id: str
+    title: str
+    chat: dict
+    updated_at: int  # timestamp in epoch
+    created_at: int  # timestamp in epoch
+    share_id: Optional[str] = None  # id of the chat to be shared
+    archived: bool
+
+
+class ChatTitleIdResponse(BaseModel):
+    id: str
+    title: str
+    updated_at: int
+    created_at: int
+
+
+class ChatTable:
+    def insert_new_chat(self, user_id: str, form_data: ChatForm) -> Optional[ChatModel]:
+        with get_db() as db:
+            id = str(uuid.uuid4())
+            chat = ChatModel(
+                **{
+                    "id": id,
+                    "user_id": user_id,
+                    "title": (
+                        form_data.chat["title"]
+                        if "title" in form_data.chat
+                        else "New Chat"
+                    ),
+                    "chat": json.dumps(form_data.chat),
+                    "created_at": int(time.time()),
+                    "updated_at": int(time.time()),
+                }
+            )
+
+            result = Chat(**chat.model_dump())
+            db.add(result)
+            db.commit()
+            db.refresh(result)
+            return ChatModel.model_validate(result) if result else None
+
+    def update_chat_by_id(self, id: str, chat: dict) -> Optional[ChatModel]:
+        try:
+            with get_db() as db:
+                chat_obj = db.get(Chat, id)
+                chat_obj.chat = json.dumps(chat)
+                chat_obj.title = chat["title"] if "title" in chat else "New Chat"
+                chat_obj.updated_at = int(time.time())
+                db.commit()
+                db.refresh(chat_obj)
+
+                return ChatModel.model_validate(chat_obj)
+        except Exception:
+            return None
+
+    def insert_shared_chat_by_chat_id(self, chat_id: str) -> Optional[ChatModel]:
+        with get_db() as db:
+            # Get the existing chat to share
+            chat = db.get(Chat, chat_id)
+            # Check if the chat is already shared
+            if chat.share_id:
+                return self.get_chat_by_id_and_user_id(chat.share_id, "shared")
+            # Create a new chat with the same data, but with a new ID
+            shared_chat = ChatModel(
+                **{
+                    "id": str(uuid.uuid4()),
+                    "user_id": f"shared-{chat_id}",
+                    "title": chat.title,
+                    "chat": chat.chat,
+                    "created_at": chat.created_at,
+                    "updated_at": int(time.time()),
+                }
+            )
+            shared_result = Chat(**shared_chat.model_dump())
+            db.add(shared_result)
+            db.commit()
+            db.refresh(shared_result)
+
+            # Update the original chat with the share_id
+            result = (
+                db.query(Chat)
+                .filter_by(id=chat_id)
+                .update({"share_id": shared_chat.id})
+            )
+            db.commit()
+            return shared_chat if (shared_result and result) else None
+
+    def update_shared_chat_by_chat_id(self, chat_id: str) -> Optional[ChatModel]:
+        try:
+            with get_db() as db:
+                print("update_shared_chat_by_id")
+                chat = db.get(Chat, chat_id)
+                print(chat)
+                chat.title = chat.title
+                chat.chat = chat.chat
+                db.commit()
+                db.refresh(chat)
+
+                return self.get_chat_by_id(chat.share_id)
+        except Exception:
+            return None
+
+    def delete_shared_chat_by_chat_id(self, chat_id: str) -> bool:
+        try:
+            with get_db() as db:
+                db.query(Chat).filter_by(user_id=f"shared-{chat_id}").delete()
+                db.commit()
+
+                return True
+        except Exception:
+            return False
+
+    def update_chat_share_id_by_id(
+        self, id: str, share_id: Optional[str]
+    ) -> Optional[ChatModel]:
+        try:
+            with get_db() as db:
+                chat = db.get(Chat, id)
+                chat.share_id = share_id
+                db.commit()
+                db.refresh(chat)
+                return ChatModel.model_validate(chat)
+        except Exception:
+            return None
+
+    def toggle_chat_archive_by_id(self, id: str) -> Optional[ChatModel]:
+        try:
+            with get_db() as db:
+                chat = db.get(Chat, id)
+                chat.archived = not chat.archived
+                db.commit()
+                db.refresh(chat)
+                return ChatModel.model_validate(chat)
+        except Exception:
+            return None
+
+    def archive_all_chats_by_user_id(self, user_id: str) -> bool:
+        try:
+            with get_db() as db:
+                db.query(Chat).filter_by(user_id=user_id).update({"archived": True})
+                db.commit()
+                return True
+        except Exception:
+            return False
+
+    def get_archived_chat_list_by_user_id(
+        self, user_id: str, skip: int = 0, limit: int = 50
+    ) -> list[ChatModel]:
+        with get_db() as db:
+            all_chats = (
+                db.query(Chat)
+                .filter_by(user_id=user_id, archived=True)
+                .order_by(Chat.updated_at.desc())
+                # .limit(limit).offset(skip)
+                .all()
+            )
+            return [ChatModel.model_validate(chat) for chat in all_chats]
+
+    def get_chat_list_by_user_id(
+        self,
+        user_id: str,
+        include_archived: bool = False,
+        skip: int = 0,
+        limit: int = 50,
+    ) -> list[ChatModel]:
+        with get_db() as db:
+            query = db.query(Chat).filter_by(user_id=user_id)
+            if not include_archived:
+                query = query.filter_by(archived=False)
+            all_chats = (
+                query.order_by(Chat.updated_at.desc())
+                # .limit(limit).offset(skip)
+                .all()
+            )
+            return [ChatModel.model_validate(chat) for chat in all_chats]
+
+    def get_chat_title_id_list_by_user_id(
+        self,
+        user_id: str,
+        include_archived: bool = False,
+        skip: Optional[int] = None,
+        limit: Optional[int] = None,
+    ) -> list[ChatTitleIdResponse]:
+        with get_db() as db:
+            query = db.query(Chat).filter_by(user_id=user_id)
+            if not include_archived:
+                query = query.filter_by(archived=False)
+
+            query = query.order_by(Chat.updated_at.desc()).with_entities(
+                Chat.id, Chat.title, Chat.updated_at, Chat.created_at
+            )
+
+            if limit:
+                query = query.limit(limit)
+            if skip:
+                query = query.offset(skip)
+
+            all_chats = query.all()
+
+            # result has to be destrctured from sqlalchemy `row` and mapped to a dict since the `ChatModel`is not the returned dataclass.
+            return [
+                ChatTitleIdResponse.model_validate(
+                    {
+                        "id": chat[0],
+                        "title": chat[1],
+                        "updated_at": chat[2],
+                        "created_at": chat[3],
+                    }
+                )
+                for chat in all_chats
+            ]
+
+    def get_chat_list_by_chat_ids(
+        self, chat_ids: list[str], skip: int = 0, limit: int = 50
+    ) -> list[ChatModel]:
+        with get_db() as db:
+            all_chats = (
+                db.query(Chat)
+                .filter(Chat.id.in_(chat_ids))
+                .filter_by(archived=False)
+                .order_by(Chat.updated_at.desc())
+                .all()
+            )
+            return [ChatModel.model_validate(chat) for chat in all_chats]
+
+    def get_chat_by_id(self, id: str) -> Optional[ChatModel]:
+        try:
+            with get_db() as db:
+                chat = db.get(Chat, id)
+                return ChatModel.model_validate(chat)
+        except Exception:
+            return None
+
+    def get_chat_by_share_id(self, id: str) -> Optional[ChatModel]:
+        try:
+            with get_db() as db:
+                chat = db.query(Chat).filter_by(share_id=id).first()
+
+                if chat:
+                    return self.get_chat_by_id(id)
+                else:
+                    return None
+        except Exception:
+            return None
+
+    def get_chat_by_id_and_user_id(self, id: str, user_id: str) -> Optional[ChatModel]:
+        try:
+            with get_db() as db:
+                chat = db.query(Chat).filter_by(id=id, user_id=user_id).first()
+                return ChatModel.model_validate(chat)
+        except Exception:
+            return None
+
+    def get_chats(self, skip: int = 0, limit: int = 50) -> list[ChatModel]:
+        with get_db() as db:
+            all_chats = (
+                db.query(Chat)
+                # .limit(limit).offset(skip)
+                .order_by(Chat.updated_at.desc())
+            )
+            return [ChatModel.model_validate(chat) for chat in all_chats]
+
+    def get_chats_by_user_id(self, user_id: str) -> list[ChatModel]:
+        with get_db() as db:
+            all_chats = (
+                db.query(Chat)
+                .filter_by(user_id=user_id)
+                .order_by(Chat.updated_at.desc())
+            )
+            return [ChatModel.model_validate(chat) for chat in all_chats]
+
+    def get_archived_chats_by_user_id(self, user_id: str) -> list[ChatModel]:
+        with get_db() as db:
+            all_chats = (
+                db.query(Chat)
+                .filter_by(user_id=user_id, archived=True)
+                .order_by(Chat.updated_at.desc())
+            )
+            return [ChatModel.model_validate(chat) for chat in all_chats]
+
+    def delete_chat_by_id(self, id: str) -> bool:
+        try:
+            with get_db() as db:
+                db.query(Chat).filter_by(id=id).delete()
+                db.commit()
+
+                return True and self.delete_shared_chat_by_chat_id(id)
+        except Exception:
+            return False
+
+    def delete_chat_by_id_and_user_id(self, id: str, user_id: str) -> bool:
+        try:
+            with get_db() as db:
+                db.query(Chat).filter_by(id=id, user_id=user_id).delete()
+                db.commit()
+
+                return True and self.delete_shared_chat_by_chat_id(id)
+        except Exception:
+            return False
+
+    def delete_chats_by_user_id(self, user_id: str) -> bool:
+        try:
+            with get_db() as db:
+                self.delete_shared_chats_by_user_id(user_id)
+
+                db.query(Chat).filter_by(user_id=user_id).delete()
+                db.commit()
+
+                return True
+        except Exception:
+            return False
+
+    def delete_shared_chats_by_user_id(self, user_id: str) -> bool:
+        try:
+            with get_db() as db:
+                chats_by_user = db.query(Chat).filter_by(user_id=user_id).all()
+                shared_chat_ids = [f"shared-{chat.id}" for chat in chats_by_user]
+
+                db.query(Chat).filter(Chat.user_id.in_(shared_chat_ids)).delete()
+                db.commit()
+
+                return True
+        except Exception:
+            return False
+
+
+Chats = ChatTable()

backend/open_webui/apps/webui/models/documents.py

@@ -0,0 +1,157 @@
+import json
+import logging
+import time
+from typing import Optional
+
+from open_webui.apps.webui.internal.db import Base, get_db
+from open_webui.env import SRC_LOG_LEVELS
+from pydantic import BaseModel, ConfigDict
+from sqlalchemy import BigInteger, Column, String, Text
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MODELS"])
+
+####################
+# Documents DB Schema
+####################
+
+
+class Document(Base):
+    __tablename__ = "document"
+
+    collection_name = Column(String, primary_key=True)
+    name = Column(String, unique=True)
+    title = Column(Text)
+    filename = Column(Text)
+    content = Column(Text, nullable=True)
+    user_id = Column(String)
+    timestamp = Column(BigInteger)
+
+
+class DocumentModel(BaseModel):
+    model_config = ConfigDict(from_attributes=True)
+
+    collection_name: str
+    name: str
+    title: str
+    filename: str
+    content: Optional[str] = None
+    user_id: str
+    timestamp: int  # timestamp in epoch
+
+
+####################
+# Forms
+####################
+
+
+class DocumentResponse(BaseModel):
+    collection_name: str
+    name: str
+    title: str
+    filename: str
+    content: Optional[dict] = None
+    user_id: str
+    timestamp: int  # timestamp in epoch
+
+
+class DocumentUpdateForm(BaseModel):
+    name: str
+    title: str
+
+
+class DocumentForm(DocumentUpdateForm):
+    collection_name: str
+    filename: str
+    content: Optional[str] = None
+
+
+class DocumentsTable:
+    def insert_new_doc(
+        self, user_id: str, form_data: DocumentForm
+    ) -> Optional[DocumentModel]:
+        with get_db() as db:
+            document = DocumentModel(
+                **{
+                    **form_data.model_dump(),
+                    "user_id": user_id,
+                    "timestamp": int(time.time()),
+                }
+            )
+
+            try:
+                result = Document(**document.model_dump())
+                db.add(result)
+                db.commit()
+                db.refresh(result)
+                if result:
+                    return DocumentModel.model_validate(result)
+                else:
+                    return None
+            except Exception:
+                return None
+
+    def get_doc_by_name(self, name: str) -> Optional[DocumentModel]:
+        try:
+            with get_db() as db:
+                document = db.query(Document).filter_by(name=name).first()
+                return DocumentModel.model_validate(document) if document else None
+        except Exception:
+            return None
+
+    def get_docs(self) -> list[DocumentModel]:
+        with get_db() as db:
+            return [
+                DocumentModel.model_validate(doc) for doc in db.query(Document).all()
+            ]
+
+    def update_doc_by_name(
+        self, name: str, form_data: DocumentUpdateForm
+    ) -> Optional[DocumentModel]:
+        try:
+            with get_db() as db:
+                db.query(Document).filter_by(name=name).update(
+                    {
+                        "title": form_data.title,
+                        "name": form_data.name,
+                        "timestamp": int(time.time()),
+                    }
+                )
+                db.commit()
+                return self.get_doc_by_name(form_data.name)
+        except Exception as e:
+            log.exception(e)
+            return None
+
+    def update_doc_content_by_name(
+        self, name: str, updated: dict
+    ) -> Optional[DocumentModel]:
+        try:
+            doc = self.get_doc_by_name(name)
+            doc_content = json.loads(doc.content if doc.content else "{}")
+            doc_content = {**doc_content, **updated}
+
+            with get_db() as db:
+                db.query(Document).filter_by(name=name).update(
+                    {
+                        "content": json.dumps(doc_content),
+                        "timestamp": int(time.time()),
+                    }
+                )
+                db.commit()
+                return self.get_doc_by_name(name)
+        except Exception as e:
+            log.exception(e)
+            return None
+
+    def delete_doc_by_name(self, name: str) -> bool:
+        try:
+            with get_db() as db:
+                db.query(Document).filter_by(name=name).delete()
+                db.commit()
+                return True
+        except Exception:
+            return False
+
+
+Documents = DocumentsTable()

backend/open_webui/apps/webui/models/files.py

@@ -0,0 +1,121 @@
+import logging
+import time
+from typing import Optional
+
+from open_webui.apps.webui.internal.db import Base, JSONField, get_db
+from open_webui.env import SRC_LOG_LEVELS
+from pydantic import BaseModel, ConfigDict
+from sqlalchemy import BigInteger, Column, String, Text
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MODELS"])
+
+####################
+# Files DB Schema
+####################
+
+
+class File(Base):
+    __tablename__ = "file"
+
+    id = Column(String, primary_key=True)
+    user_id = Column(String)
+    filename = Column(Text)
+    meta = Column(JSONField)
+    created_at = Column(BigInteger)
+
+
+class FileModel(BaseModel):
+    id: str
+    user_id: str
+    filename: str
+    meta: dict
+    created_at: int  # timestamp in epoch
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+####################
+# Forms
+####################
+
+
+class FileModelResponse(BaseModel):
+    id: str
+    user_id: str
+    filename: str
+    meta: dict
+    created_at: int  # timestamp in epoch
+
+
+class FileForm(BaseModel):
+    id: str
+    filename: str
+    meta: dict = {}
+
+
+class FilesTable:
+    def insert_new_file(self, user_id: str, form_data: FileForm) -> Optional[FileModel]:
+        with get_db() as db:
+            file = FileModel(
+                **{
+                    **form_data.model_dump(),
+                    "user_id": user_id,
+                    "created_at": int(time.time()),
+                }
+            )
+
+            try:
+                result = File(**file.model_dump())
+                db.add(result)
+                db.commit()
+                db.refresh(result)
+                if result:
+                    return FileModel.model_validate(result)
+                else:
+                    return None
+            except Exception as e:
+                print(f"Error creating tool: {e}")
+                return None
+
+    def get_file_by_id(self, id: str) -> Optional[FileModel]:
+        with get_db() as db:
+            try:
+                file = db.get(File, id)
+                return FileModel.model_validate(file)
+            except Exception:
+                return None
+
+    def get_files(self) -> list[FileModel]:
+        with get_db() as db:
+            return [FileModel.model_validate(file) for file in db.query(File).all()]
+
+    def get_files_by_user_id(self, user_id: str) -> list[FileModel]:
+        with get_db() as db:
+            return [
+                FileModel.model_validate(file)
+                for file in db.query(File).filter_by(user_id=user_id).all()
+            ]
+
+    def delete_file_by_id(self, id: str) -> bool:
+        with get_db() as db:
+            try:
+                db.query(File).filter_by(id=id).delete()
+                db.commit()
+
+                return True
+            except Exception:
+                return False
+
+    def delete_all_files(self) -> bool:
+        with get_db() as db:
+            try:
+                db.query(File).delete()
+                db.commit()
+
+                return True
+            except Exception:
+                return False
+
+
+Files = FilesTable()

backend/open_webui/apps/webui/models/functions.py

@@ -0,0 +1,270 @@
+import logging
+import time
+from typing import Optional
+
+from open_webui.apps.webui.internal.db import Base, JSONField, get_db
+from open_webui.apps.webui.models.users import Users
+from open_webui.env import SRC_LOG_LEVELS
+from pydantic import BaseModel, ConfigDict
+from sqlalchemy import BigInteger, Boolean, Column, String, Text
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MODELS"])
+
+####################
+# Functions DB Schema
+####################
+
+
+class Function(Base):
+    __tablename__ = "function"
+
+    id = Column(String, primary_key=True)
+    user_id = Column(String)
+    name = Column(Text)
+    type = Column(Text)
+    content = Column(Text)
+    meta = Column(JSONField)
+    valves = Column(JSONField)
+    is_active = Column(Boolean)
+    is_global = Column(Boolean)
+    updated_at = Column(BigInteger)
+    created_at = Column(BigInteger)
+
+
+class FunctionMeta(BaseModel):
+    description: Optional[str] = None
+    manifest: Optional[dict] = {}
+
+
+class FunctionModel(BaseModel):
+    id: str
+    user_id: str
+    name: str
+    type: str
+    content: str
+    meta: FunctionMeta
+    is_active: bool = False
+    is_global: bool = False
+    updated_at: int  # timestamp in epoch
+    created_at: int  # timestamp in epoch
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+####################
+# Forms
+####################
+
+
+class FunctionResponse(BaseModel):
+    id: str
+    user_id: str
+    type: str
+    name: str
+    meta: FunctionMeta
+    is_active: bool
+    is_global: bool
+    updated_at: int  # timestamp in epoch
+    created_at: int  # timestamp in epoch
+
+
+class FunctionForm(BaseModel):
+    id: str
+    name: str
+    content: str
+    meta: FunctionMeta
+
+
+class FunctionValves(BaseModel):
+    valves: Optional[dict] = None
+
+
+class FunctionsTable:
+    def insert_new_function(
+        self, user_id: str, type: str, form_data: FunctionForm
+    ) -> Optional[FunctionModel]:
+        function = FunctionModel(
+            **{
+                **form_data.model_dump(),
+                "user_id": user_id,
+                "type": type,
+                "updated_at": int(time.time()),
+                "created_at": int(time.time()),
+            }
+        )
+
+        try:
+            with get_db() as db:
+                result = Function(**function.model_dump())
+                db.add(result)
+                db.commit()
+                db.refresh(result)
+                if result:
+                    return FunctionModel.model_validate(result)
+                else:
+                    return None
+        except Exception as e:
+            print(f"Error creating tool: {e}")
+            return None
+
+    def get_function_by_id(self, id: str) -> Optional[FunctionModel]:
+        try:
+            with get_db() as db:
+                function = db.get(Function, id)
+                return FunctionModel.model_validate(function)
+        except Exception:
+            return None
+
+    def get_functions(self, active_only=False) -> list[FunctionModel]:
+        with get_db() as db:
+            if active_only:
+                return [
+                    FunctionModel.model_validate(function)
+                    for function in db.query(Function).filter_by(is_active=True).all()
+                ]
+            else:
+                return [
+                    FunctionModel.model_validate(function)
+                    for function in db.query(Function).all()
+                ]
+
+    def get_functions_by_type(
+        self, type: str, active_only=False
+    ) -> list[FunctionModel]:
+        with get_db() as db:
+            if active_only:
+                return [
+                    FunctionModel.model_validate(function)
+                    for function in db.query(Function)
+                    .filter_by(type=type, is_active=True)
+                    .all()
+                ]
+            else:
+                return [
+                    FunctionModel.model_validate(function)
+                    for function in db.query(Function).filter_by(type=type).all()
+                ]
+
+    def get_global_filter_functions(self) -> list[FunctionModel]:
+        with get_db() as db:
+            return [
+                FunctionModel.model_validate(function)
+                for function in db.query(Function)
+                .filter_by(type="filter", is_active=True, is_global=True)
+                .all()
+            ]
+
+    def get_global_action_functions(self) -> list[FunctionModel]:
+        with get_db() as db:
+            return [
+                FunctionModel.model_validate(function)
+                for function in db.query(Function)
+                .filter_by(type="action", is_active=True, is_global=True)
+                .all()
+            ]
+
+    def get_function_valves_by_id(self, id: str) -> Optional[dict]:
+        with get_db() as db:
+            try:
+                function = db.get(Function, id)
+                return function.valves if function.valves else {}
+            except Exception as e:
+                print(f"An error occurred: {e}")
+                return None
+
+    def update_function_valves_by_id(
+        self, id: str, valves: dict
+    ) -> Optional[FunctionValves]:
+        with get_db() as db:
+            try:
+                function = db.get(Function, id)
+                function.valves = valves
+                function.updated_at = int(time.time())
+                db.commit()
+                db.refresh(function)
+                return self.get_function_by_id(id)
+            except Exception:
+                return None
+
+    def get_user_valves_by_id_and_user_id(
+        self, id: str, user_id: str
+    ) -> Optional[dict]:
+        try:
+            user = Users.get_user_by_id(user_id)
+            user_settings = user.settings.model_dump() if user.settings else {}
+
+            # Check if user has "functions" and "valves" settings
+            if "functions" not in user_settings:
+                user_settings["functions"] = {}
+            if "valves" not in user_settings["functions"]:
+                user_settings["functions"]["valves"] = {}
+
+            return user_settings["functions"]["valves"].get(id, {})
+        except Exception as e:
+            print(f"An error occurred: {e}")
+            return None
+
+    def update_user_valves_by_id_and_user_id(
+        self, id: str, user_id: str, valves: dict
+    ) -> Optional[dict]:
+        try:
+            user = Users.get_user_by_id(user_id)
+            user_settings = user.settings.model_dump() if user.settings else {}
+
+            # Check if user has "functions" and "valves" settings
+            if "functions" not in user_settings:
+                user_settings["functions"] = {}
+            if "valves" not in user_settings["functions"]:
+                user_settings["functions"]["valves"] = {}
+
+            user_settings["functions"]["valves"][id] = valves
+
+            # Update the user settings in the database
+            Users.update_user_by_id(user_id, {"settings": user_settings})
+
+            return user_settings["functions"]["valves"][id]
+        except Exception as e:
+            print(f"An error occurred: {e}")
+            return None
+
+    def update_function_by_id(self, id: str, updated: dict) -> Optional[FunctionModel]:
+        with get_db() as db:
+            try:
+                db.query(Function).filter_by(id=id).update(
+                    {
+                        **updated,
+                        "updated_at": int(time.time()),
+                    }
+                )
+                db.commit()
+                return self.get_function_by_id(id)
+            except Exception:
+                return None
+
+    def deactivate_all_functions(self) -> Optional[bool]:
+        with get_db() as db:
+            try:
+                db.query(Function).update(
+                    {
+                        "is_active": False,
+                        "updated_at": int(time.time()),
+                    }
+                )
+                db.commit()
+                return True
+            except Exception:
+                return None
+
+    def delete_function_by_id(self, id: str) -> bool:
+        with get_db() as db:
+            try:
+                db.query(Function).filter_by(id=id).delete()
+                db.commit()
+
+                return True
+            except Exception:
+                return False
+
+
+Functions = FunctionsTable()

backend/open_webui/apps/webui/models/memories.py

@@ -0,0 +1,137 @@
+import time
+import uuid
+from typing import Optional
+
+from open_webui.apps.webui.internal.db import Base, get_db
+from pydantic import BaseModel, ConfigDict
+from sqlalchemy import BigInteger, Column, String, Text
+
+####################
+# Memory DB Schema
+####################
+
+
+class Memory(Base):
+    __tablename__ = "memory"
+
+    id = Column(String, primary_key=True)
+    user_id = Column(String)
+    content = Column(Text)
+    updated_at = Column(BigInteger)
+    created_at = Column(BigInteger)
+
+
+class MemoryModel(BaseModel):
+    id: str
+    user_id: str
+    content: str
+    updated_at: int  # timestamp in epoch
+    created_at: int  # timestamp in epoch
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+####################
+# Forms
+####################
+
+
+class MemoriesTable:
+    def insert_new_memory(
+        self,
+        user_id: str,
+        content: str,
+    ) -> Optional[MemoryModel]:
+        with get_db() as db:
+            id = str(uuid.uuid4())
+
+            memory = MemoryModel(
+                **{
+                    "id": id,
+                    "user_id": user_id,
+                    "content": content,
+                    "created_at": int(time.time()),
+                    "updated_at": int(time.time()),
+                }
+            )
+            result = Memory(**memory.model_dump())
+            db.add(result)
+            db.commit()
+            db.refresh(result)
+            if result:
+                return MemoryModel.model_validate(result)
+            else:
+                return None
+
+    def update_memory_by_id(
+        self,
+        id: str,
+        content: str,
+    ) -> Optional[MemoryModel]:
+        with get_db() as db:
+            try:
+                db.query(Memory).filter_by(id=id).update(
+                    {"content": content, "updated_at": int(time.time())}
+                )
+                db.commit()
+                return self.get_memory_by_id(id)
+            except Exception:
+                return None
+
+    def get_memories(self) -> list[MemoryModel]:
+        with get_db() as db:
+            try:
+                memories = db.query(Memory).all()
+                return [MemoryModel.model_validate(memory) for memory in memories]
+            except Exception:
+                return None
+
+    def get_memories_by_user_id(self, user_id: str) -> list[MemoryModel]:
+        with get_db() as db:
+            try:
+                memories = db.query(Memory).filter_by(user_id=user_id).all()
+                return [MemoryModel.model_validate(memory) for memory in memories]
+            except Exception:
+                return None
+
+    def get_memory_by_id(self, id: str) -> Optional[MemoryModel]:
+        with get_db() as db:
+            try:
+                memory = db.get(Memory, id)
+                return MemoryModel.model_validate(memory)
+            except Exception:
+                return None
+
+    def delete_memory_by_id(self, id: str) -> bool:
+        with get_db() as db:
+            try:
+                db.query(Memory).filter_by(id=id).delete()
+                db.commit()
+
+                return True
+
+            except Exception:
+                return False
+
+    def delete_memories_by_user_id(self, user_id: str) -> bool:
+        with get_db() as db:
+            try:
+                db.query(Memory).filter_by(user_id=user_id).delete()
+                db.commit()
+
+                return True
+            except Exception:
+                return False
+
+    def delete_memory_by_id_and_user_id(self, id: str, user_id: str) -> bool:
+        with get_db() as db:
+            try:
+                db.query(Memory).filter_by(id=id, user_id=user_id).delete()
+                db.commit()
+
+                return True
+            except Exception:
+                return False
+
+
+Memories = MemoriesTable()

backend/open_webui/apps/webui/models/models.py

@@ -1,19 +1,11 @@
-import json
 import logging
+import time
 from typing import Optional
 
-import peewee as pw
-from peewee import *
-
-from playhouse.shortcuts import model_to_dict
+from open_webui.apps.webui.internal.db import Base, JSONField, get_db
+from open_webui.env import SRC_LOG_LEVELS
 from pydantic import BaseModel, ConfigDict
-
-from apps.webui.internal.db import DB, JSONField
-
-from typing import List, Union, Optional
-from config import SRC_LOG_LEVELS
-
-import time
+from sqlalchemy import BigInteger, Column, Text
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])
@@ -32,7 +24,7 @@ class ModelParams(BaseModel):
 
 # ModelMeta is a model for the data stored in the meta field of the Model table
 class ModelMeta(BaseModel):
-    profile_image_url: Optional[str] = "/favicon.png"
+    profile_image_url: Optional[str] = "/static/favicon.png"
 
     description: Optional[str] = None
     """
@@ -46,38 +38,37 @@ class ModelMeta(BaseModel):
     pass
 
 
-class Model(pw.Model):
-    id = pw.TextField(unique=True)
+class Model(Base):
+    __tablename__ = "model"
+
+    id = Column(Text, primary_key=True)
     """
         The model's id as used in the API. If set to an existing model, it will override the model.
     """
-    user_id = pw.TextField()
+    user_id = Column(Text)
 
-    base_model_id = pw.TextField(null=True)
+    base_model_id = Column(Text, nullable=True)
     """
         An optional pointer to the actual model that should be used when proxying requests.
     """
 
-    name = pw.TextField()
+    name = Column(Text)
     """
         The human-readable display name of the model.
     """
 
-    params = JSONField()
+    params = Column(JSONField)
     """
         Holds a JSON encoded blob of parameters, see `ModelParams`.
     """
 
-    meta = JSONField()
+    meta = Column(JSONField)
     """
         Holds a JSON encoded blob of metadata, see `ModelMeta`.
     """
 
-    updated_at = BigIntegerField()
-    created_at = BigIntegerField()
-
-    class Meta:
-        database = DB
+    updated_at = Column(BigInteger)
+    created_at = Column(BigInteger)
 
 
 class ModelModel(BaseModel):
@@ -92,6 +83,8 @@ class ModelModel(BaseModel):
     updated_at: int  # timestamp in epoch
     created_at: int  # timestamp in epoch
 
+    model_config = ConfigDict(from_attributes=True)
+
 
 ####################
 # Forms
@@ -115,13 +108,6 @@ class ModelForm(BaseModel):
 
 
 class ModelsTable:
-    def __init__(
-        self,
-        db: pw.SqliteDatabase | pw.PostgresqlDatabase,
-    ):
-        self.db = db
-        self.db.create_tables([Model])
-
     def insert_new_model(
         self, form_data: ModelForm, user_id: str
     ) -> Optional[ModelModel]:
@@ -134,34 +120,46 @@ class ModelsTable:
             }
         )
         try:
-            result = Model.create(**model.model_dump())
+            with get_db() as db:
+                result = Model(**model.model_dump())
+                db.add(result)
+                db.commit()
+                db.refresh(result)
 
-            if result:
-                return model
-            else:
-                return None
+                if result:
+                    return ModelModel.model_validate(result)
+                else:
+                    return None
         except Exception as e:
             print(e)
             return None
 
-    def get_all_models(self) -> List[ModelModel]:
-        return [ModelModel(**model_to_dict(model)) for model in Model.select()]
+    def get_all_models(self) -> list[ModelModel]:
+        with get_db() as db:
+            return [ModelModel.model_validate(model) for model in db.query(Model).all()]
 
     def get_model_by_id(self, id: str) -> Optional[ModelModel]:
         try:
-            model = Model.get(Model.id == id)
-            return ModelModel(**model_to_dict(model))
-        except:
+            with get_db() as db:
+                model = db.get(Model, id)
+                return ModelModel.model_validate(model)
+        except Exception:
             return None
 
     def update_model_by_id(self, id: str, model: ModelForm) -> Optional[ModelModel]:
         try:
-            # update only the fields that are present in the model
-            query = Model.update(**model.model_dump()).where(Model.id == id)
-            query.execute()
+            with get_db() as db:
+                # update only the fields that are present in the model
+                result = (
+                    db.query(Model)
+                    .filter_by(id=id)
+                    .update(model.model_dump(exclude={"id"}, exclude_none=True))
+                )
+                db.commit()
 
-            model = Model.get(Model.id == id)
-            return ModelModel(**model_to_dict(model))
+                model = db.get(Model, id)
+                db.refresh(model)
+                return ModelModel.model_validate(model)
         except Exception as e:
             print(e)
 
@@ -169,11 +167,13 @@ class ModelsTable:
 
     def delete_model_by_id(self, id: str) -> bool:
         try:
-            query = Model.delete().where(Model.id == id)
-            query.execute()
-            return True
-        except:
+            with get_db() as db:
+                db.query(Model).filter_by(id=id).delete()
+                db.commit()
+
+                return True
+        except Exception:
             return False
 
 
-Models = ModelsTable(DB)
+Models = ModelsTable()

backend/open_webui/apps/webui/models/prompts.py

@@ -0,0 +1,110 @@
+import time
+from typing import Optional
+
+from open_webui.apps.webui.internal.db import Base, get_db
+from pydantic import BaseModel, ConfigDict
+from sqlalchemy import BigInteger, Column, String, Text
+
+####################
+# Prompts DB Schema
+####################
+
+
+class Prompt(Base):
+    __tablename__ = "prompt"
+
+    command = Column(String, primary_key=True)
+    user_id = Column(String)
+    title = Column(Text)
+    content = Column(Text)
+    timestamp = Column(BigInteger)
+
+
+class PromptModel(BaseModel):
+    command: str
+    user_id: str
+    title: str
+    content: str
+    timestamp: int  # timestamp in epoch
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+####################
+# Forms
+####################
+
+
+class PromptForm(BaseModel):
+    command: str
+    title: str
+    content: str
+
+
+class PromptsTable:
+    def insert_new_prompt(
+        self, user_id: str, form_data: PromptForm
+    ) -> Optional[PromptModel]:
+        prompt = PromptModel(
+            **{
+                "user_id": user_id,
+                "command": form_data.command,
+                "title": form_data.title,
+                "content": form_data.content,
+                "timestamp": int(time.time()),
+            }
+        )
+
+        try:
+            with get_db() as db:
+                result = Prompt(**prompt.dict())
+                db.add(result)
+                db.commit()
+                db.refresh(result)
+                if result:
+                    return PromptModel.model_validate(result)
+                else:
+                    return None
+        except Exception:
+            return None
+
+    def get_prompt_by_command(self, command: str) -> Optional[PromptModel]:
+        try:
+            with get_db() as db:
+                prompt = db.query(Prompt).filter_by(command=command).first()
+                return PromptModel.model_validate(prompt)
+        except Exception:
+            return None
+
+    def get_prompts(self) -> list[PromptModel]:
+        with get_db() as db:
+            return [
+                PromptModel.model_validate(prompt) for prompt in db.query(Prompt).all()
+            ]
+
+    def update_prompt_by_command(
+        self, command: str, form_data: PromptForm
+    ) -> Optional[PromptModel]:
+        try:
+            with get_db() as db:
+                prompt = db.query(Prompt).filter_by(command=command).first()
+                prompt.title = form_data.title
+                prompt.content = form_data.content
+                prompt.timestamp = int(time.time())
+                db.commit()
+                return PromptModel.model_validate(prompt)
+        except Exception:
+            return None
+
+    def delete_prompt_by_command(self, command: str) -> bool:
+        try:
+            with get_db() as db:
+                db.query(Prompt).filter_by(command=command).delete()
+                db.commit()
+
+                return True
+        except Exception:
+            return False
+
+
+Prompts = PromptsTable()

backend/open_webui/apps/webui/models/tags.py

@@ -0,0 +1,262 @@
+import logging
+import time
+import uuid
+from typing import Optional
+
+from open_webui.apps.webui.internal.db import Base, get_db
+from open_webui.env import SRC_LOG_LEVELS
+from pydantic import BaseModel, ConfigDict
+from sqlalchemy import BigInteger, Column, String, Text
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MODELS"])
+
+####################
+# Tag DB Schema
+####################
+
+
+class Tag(Base):
+    __tablename__ = "tag"
+
+    id = Column(String, primary_key=True)
+    name = Column(String)
+    user_id = Column(String)
+    data = Column(Text, nullable=True)
+
+
+class ChatIdTag(Base):
+    __tablename__ = "chatidtag"
+
+    id = Column(String, primary_key=True)
+    tag_name = Column(String)
+    chat_id = Column(String)
+    user_id = Column(String)
+    timestamp = Column(BigInteger)
+
+
+class TagModel(BaseModel):
+    id: str
+    name: str
+    user_id: str
+    data: Optional[str] = None
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+class ChatIdTagModel(BaseModel):
+    id: str
+    tag_name: str
+    chat_id: str
+    user_id: str
+    timestamp: int
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+####################
+# Forms
+####################
+
+
+class ChatIdTagForm(BaseModel):
+    tag_name: str
+    chat_id: str
+
+
+class TagChatIdsResponse(BaseModel):
+    chat_ids: list[str]
+
+
+class ChatTagsResponse(BaseModel):
+    tags: list[str]
+
+
+class TagTable:
+    def insert_new_tag(self, name: str, user_id: str) -> Optional[TagModel]:
+        with get_db() as db:
+            id = str(uuid.uuid4())
+            tag = TagModel(**{"id": id, "user_id": user_id, "name": name})
+            try:
+                result = Tag(**tag.model_dump())
+                db.add(result)
+                db.commit()
+                db.refresh(result)
+                if result:
+                    return TagModel.model_validate(result)
+                else:
+                    return None
+            except Exception:
+                return None
+
+    def get_tag_by_name_and_user_id(
+        self, name: str, user_id: str
+    ) -> Optional[TagModel]:
+        try:
+            with get_db() as db:
+                tag = db.query(Tag).filter_by(name=name, user_id=user_id).first()
+                return TagModel.model_validate(tag)
+        except Exception:
+            return None
+
+    def add_tag_to_chat(
+        self, user_id: str, form_data: ChatIdTagForm
+    ) -> Optional[ChatIdTagModel]:
+        tag = self.get_tag_by_name_and_user_id(form_data.tag_name, user_id)
+        if tag is None:
+            tag = self.insert_new_tag(form_data.tag_name, user_id)
+
+        id = str(uuid.uuid4())
+        chatIdTag = ChatIdTagModel(
+            **{
+                "id": id,
+                "user_id": user_id,
+                "chat_id": form_data.chat_id,
+                "tag_name": tag.name,
+                "timestamp": int(time.time()),
+            }
+        )
+        try:
+            with get_db() as db:
+                result = ChatIdTag(**chatIdTag.model_dump())
+                db.add(result)
+                db.commit()
+                db.refresh(result)
+                if result:
+                    return ChatIdTagModel.model_validate(result)
+                else:
+                    return None
+        except Exception:
+            return None
+
+    def get_tags_by_user_id(self, user_id: str) -> list[TagModel]:
+        with get_db() as db:
+            tag_names = [
+                chat_id_tag.tag_name
+                for chat_id_tag in (
+                    db.query(ChatIdTag)
+                    .filter_by(user_id=user_id)
+                    .order_by(ChatIdTag.timestamp.desc())
+                    .all()
+                )
+            ]
+
+            return [
+                TagModel.model_validate(tag)
+                for tag in (
+                    db.query(Tag)
+                    .filter_by(user_id=user_id)
+                    .filter(Tag.name.in_(tag_names))
+                    .all()
+                )
+            ]
+
+    def get_tags_by_chat_id_and_user_id(
+        self, chat_id: str, user_id: str
+    ) -> list[TagModel]:
+        with get_db() as db:
+            tag_names = [
+                chat_id_tag.tag_name
+                for chat_id_tag in (
+                    db.query(ChatIdTag)
+                    .filter_by(user_id=user_id, chat_id=chat_id)
+                    .order_by(ChatIdTag.timestamp.desc())
+                    .all()
+                )
+            ]
+
+            return [
+                TagModel.model_validate(tag)
+                for tag in (
+                    db.query(Tag)
+                    .filter_by(user_id=user_id)
+                    .filter(Tag.name.in_(tag_names))
+                    .all()
+                )
+            ]
+
+    def get_chat_ids_by_tag_name_and_user_id(
+        self, tag_name: str, user_id: str
+    ) -> list[ChatIdTagModel]:
+        with get_db() as db:
+            return [
+                ChatIdTagModel.model_validate(chat_id_tag)
+                for chat_id_tag in (
+                    db.query(ChatIdTag)
+                    .filter_by(user_id=user_id, tag_name=tag_name)
+                    .order_by(ChatIdTag.timestamp.desc())
+                    .all()
+                )
+            ]
+
+    def count_chat_ids_by_tag_name_and_user_id(
+        self, tag_name: str, user_id: str
+    ) -> int:
+        with get_db() as db:
+            return (
+                db.query(ChatIdTag)
+                .filter_by(tag_name=tag_name, user_id=user_id)
+                .count()
+            )
+
+    def delete_tag_by_tag_name_and_user_id(self, tag_name: str, user_id: str) -> bool:
+        try:
+            with get_db() as db:
+                res = (
+                    db.query(ChatIdTag)
+                    .filter_by(tag_name=tag_name, user_id=user_id)
+                    .delete()
+                )
+                log.debug(f"res: {res}")
+                db.commit()
+
+                tag_count = self.count_chat_ids_by_tag_name_and_user_id(
+                    tag_name, user_id
+                )
+                if tag_count == 0:
+                    # Remove tag item from Tag col as well
+                    db.query(Tag).filter_by(name=tag_name, user_id=user_id).delete()
+                    db.commit()
+                return True
+        except Exception as e:
+            log.error(f"delete_tag: {e}")
+            return False
+
+    def delete_tag_by_tag_name_and_chat_id_and_user_id(
+        self, tag_name: str, chat_id: str, user_id: str
+    ) -> bool:
+        try:
+            with get_db() as db:
+                res = (
+                    db.query(ChatIdTag)
+                    .filter_by(tag_name=tag_name, chat_id=chat_id, user_id=user_id)
+                    .delete()
+                )
+                log.debug(f"res: {res}")
+                db.commit()
+
+                tag_count = self.count_chat_ids_by_tag_name_and_user_id(
+                    tag_name, user_id
+                )
+                if tag_count == 0:
+                    # Remove tag item from Tag col as well
+                    db.query(Tag).filter_by(name=tag_name, user_id=user_id).delete()
+                    db.commit()
+
+                return True
+        except Exception as e:
+            log.error(f"delete_tag: {e}")
+            return False
+
+    def delete_tags_by_chat_id_and_user_id(self, chat_id: str, user_id: str) -> bool:
+        tags = self.get_tags_by_chat_id_and_user_id(chat_id, user_id)
+
+        for tag in tags:
+            self.delete_tag_by_tag_name_and_chat_id_and_user_id(
+                tag.tag_name, chat_id, user_id
+            )
+
+        return True
+
+
+Tags = TagTable()

backend/open_webui/apps/webui/models/tools.py

@@ -0,0 +1,202 @@
+import logging
+import time
+from typing import Optional
+
+from open_webui.apps.webui.internal.db import Base, JSONField, get_db
+from open_webui.apps.webui.models.users import Users
+from open_webui.env import SRC_LOG_LEVELS
+from pydantic import BaseModel, ConfigDict
+from sqlalchemy import BigInteger, Column, String, Text
+
+log = logging.getLogger(__name__)
+log.setLevel(SRC_LOG_LEVELS["MODELS"])
+
+####################
+# Tools DB Schema
+####################
+
+
+class Tool(Base):
+    __tablename__ = "tool"
+
+    id = Column(String, primary_key=True)
+    user_id = Column(String)
+    name = Column(Text)
+    content = Column(Text)
+    specs = Column(JSONField)
+    meta = Column(JSONField)
+    valves = Column(JSONField)
+    updated_at = Column(BigInteger)
+    created_at = Column(BigInteger)
+
+
+class ToolMeta(BaseModel):
+    description: Optional[str] = None
+    manifest: Optional[dict] = {}
+
+
+class ToolModel(BaseModel):
+    id: str
+    user_id: str
+    name: str
+    content: str
+    specs: list[dict]
+    meta: ToolMeta
+    updated_at: int  # timestamp in epoch
+    created_at: int  # timestamp in epoch
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+####################
+# Forms
+####################
+
+
+class ToolResponse(BaseModel):
+    id: str
+    user_id: str
+    name: str
+    meta: ToolMeta
+    updated_at: int  # timestamp in epoch
+    created_at: int  # timestamp in epoch
+
+
+class ToolForm(BaseModel):
+    id: str
+    name: str
+    content: str
+    meta: ToolMeta
+
+
+class ToolValves(BaseModel):
+    valves: Optional[dict] = None
+
+
+class ToolsTable:
+    def insert_new_tool(
+        self, user_id: str, form_data: ToolForm, specs: list[dict]
+    ) -> Optional[ToolModel]:
+        with get_db() as db:
+            tool = ToolModel(
+                **{
+                    **form_data.model_dump(),
+                    "specs": specs,
+                    "user_id": user_id,
+                    "updated_at": int(time.time()),
+                    "created_at": int(time.time()),
+                }
+            )
+
+            try:
+                result = Tool(**tool.model_dump())
+                db.add(result)
+                db.commit()
+                db.refresh(result)
+                if result:
+                    return ToolModel.model_validate(result)
+                else:
+                    return None
+            except Exception as e:
+                print(f"Error creating tool: {e}")
+                return None
+
+    def get_tool_by_id(self, id: str) -> Optional[ToolModel]:
+        try:
+            with get_db() as db:
+                tool = db.get(Tool, id)
+                return ToolModel.model_validate(tool)
+        except Exception:
+            return None
+
+    def get_tools(self) -> list[ToolModel]:
+        with get_db() as db:
+            return [ToolModel.model_validate(tool) for tool in db.query(Tool).all()]
+
+    def get_tool_valves_by_id(self, id: str) -> Optional[dict]:
+        try:
+            with get_db() as db:
+                tool = db.get(Tool, id)
+                return tool.valves if tool.valves else {}
+        except Exception as e:
+            print(f"An error occurred: {e}")
+            return None
+
+    def update_tool_valves_by_id(self, id: str, valves: dict) -> Optional[ToolValves]:
+        try:
+            with get_db() as db:
+                db.query(Tool).filter_by(id=id).update(
+                    {"valves": valves, "updated_at": int(time.time())}
+                )
+                db.commit()
+                return self.get_tool_by_id(id)
+        except Exception:
+            return None
+
+    def get_user_valves_by_id_and_user_id(
+        self, id: str, user_id: str
+    ) -> Optional[dict]:
+        try:
+            user = Users.get_user_by_id(user_id)
+            user_settings = user.settings.model_dump() if user.settings else {}
+
+            # Check if user has "tools" and "valves" settings
+            if "tools" not in user_settings:
+                user_settings["tools"] = {}
+            if "valves" not in user_settings["tools"]:
+                user_settings["tools"]["valves"] = {}
+
+            return user_settings["tools"]["valves"].get(id, {})
+        except Exception as e:
+            print(f"An error occurred: {e}")
+            return None
+
+    def update_user_valves_by_id_and_user_id(
+        self, id: str, user_id: str, valves: dict
+    ) -> Optional[dict]:
+        try:
+            user = Users.get_user_by_id(user_id)
+            user_settings = user.settings.model_dump() if user.settings else {}
+
+            # Check if user has "tools" and "valves" settings
+            if "tools" not in user_settings:
+                user_settings["tools"] = {}
+            if "valves" not in user_settings["tools"]:
+                user_settings["tools"]["valves"] = {}
+
+            user_settings["tools"]["valves"][id] = valves
+
+            # Update the user settings in the database
+            Users.update_user_by_id(user_id, {"settings": user_settings})
+
+            return user_settings["tools"]["valves"][id]
+        except Exception as e:
+            print(f"An error occurred: {e}")
+            return None
+
+    def update_tool_by_id(self, id: str, updated: dict) -> Optional[ToolModel]:
+        try:
+            with get_db() as db:
+                db.query(Tool).filter_by(id=id).update(
+                    {**updated, "updated_at": int(time.time())}
+                )
+                db.commit()
+
+                tool = db.query(Tool).get(id)
+                db.refresh(tool)
+                return ToolModel.model_validate(tool)
+        except Exception:
+            return None
+
+    def delete_tool_by_id(self, id: str) -> bool:
+        try:
+            with get_db() as db:
+                db.query(Tool).filter_by(id=id).delete()
+                db.commit()
+
+                return True
+        except Exception:
+            return False
+
+
+Tools = ToolsTable()

backend/open_webui/apps/webui/models/users.py

@@ -0,0 +1,261 @@
+import time
+from typing import Optional
+
+from open_webui.apps.webui.internal.db import Base, JSONField, get_db
+from open_webui.apps.webui.models.chats import Chats
+from pydantic import BaseModel, ConfigDict
+from sqlalchemy import BigInteger, Column, String, Text
+
+####################
+# User DB Schema
+####################
+
+
+class User(Base):
+    __tablename__ = "user"
+
+    id = Column(String, primary_key=True)
+    name = Column(String)
+    email = Column(String)
+    role = Column(String)
+    profile_image_url = Column(Text)
+
+    last_active_at = Column(BigInteger)
+    updated_at = Column(BigInteger)
+    created_at = Column(BigInteger)
+
+    api_key = Column(String, nullable=True, unique=True)
+    settings = Column(JSONField, nullable=True)
+    info = Column(JSONField, nullable=True)
+
+    oauth_sub = Column(Text, unique=True)
+
+
+class UserSettings(BaseModel):
+    ui: Optional[dict] = {}
+    model_config = ConfigDict(extra="allow")
+    pass
+
+
+class UserModel(BaseModel):
+    id: str
+    name: str
+    email: str
+    role: str = "pending"
+    profile_image_url: str
+
+    last_active_at: int  # timestamp in epoch
+    updated_at: int  # timestamp in epoch
+    created_at: int  # timestamp in epoch
+
+    api_key: Optional[str] = None
+    settings: Optional[UserSettings] = None
+    info: Optional[dict] = None
+
+    oauth_sub: Optional[str] = None
+
+    model_config = ConfigDict(from_attributes=True)
+
+
+####################
+# Forms
+####################
+
+
+class UserRoleUpdateForm(BaseModel):
+    id: str
+    role: str
+
+
+class UserUpdateForm(BaseModel):
+    name: str
+    email: str
+    profile_image_url: str
+    password: Optional[str] = None
+
+
+class UsersTable:
+    def insert_new_user(
+        self,
+        id: str,
+        name: str,
+        email: str,
+        profile_image_url: str = "/user.png",
+        role: str = "pending",
+        oauth_sub: Optional[str] = None,
+    ) -> Optional[UserModel]:
+        with get_db() as db:
+            user = UserModel(
+                **{
+                    "id": id,
+                    "name": name,
+                    "email": email,
+                    "role": role,
+                    "profile_image_url": profile_image_url,
+                    "last_active_at": int(time.time()),
+                    "created_at": int(time.time()),
+                    "updated_at": int(time.time()),
+                    "oauth_sub": oauth_sub,
+                }
+            )
+            result = User(**user.model_dump())
+            db.add(result)
+            db.commit()
+            db.refresh(result)
+            if result:
+                return user
+            else:
+                return None
+
+    def get_user_by_id(self, id: str) -> Optional[UserModel]:
+        try:
+            with get_db() as db:
+                user = db.query(User).filter_by(id=id).first()
+                return UserModel.model_validate(user)
+        except Exception:
+            return None
+
+    def get_user_by_api_key(self, api_key: str) -> Optional[UserModel]:
+        try:
+            with get_db() as db:
+                user = db.query(User).filter_by(api_key=api_key).first()
+                return UserModel.model_validate(user)
+        except Exception:
+            return None
+
+    def get_user_by_email(self, email: str) -> Optional[UserModel]:
+        try:
+            with get_db() as db:
+                user = db.query(User).filter_by(email=email).first()
+                return UserModel.model_validate(user)
+        except Exception:
+            return None
+
+    def get_user_by_oauth_sub(self, sub: str) -> Optional[UserModel]:
+        try:
+            with get_db() as db:
+                user = db.query(User).filter_by(oauth_sub=sub).first()
+                return UserModel.model_validate(user)
+        except Exception:
+            return None
+
+    def get_users(self, skip: int = 0, limit: int = 50) -> list[UserModel]:
+        with get_db() as db:
+            users = (
+                db.query(User)
+                # .offset(skip).limit(limit)
+                .all()
+            )
+            return [UserModel.model_validate(user) for user in users]
+
+    def get_num_users(self) -> Optional[int]:
+        with get_db() as db:
+            return db.query(User).count()
+
+    def get_first_user(self) -> UserModel:
+        try:
+            with get_db() as db:
+                user = db.query(User).order_by(User.created_at).first()
+                return UserModel.model_validate(user)
+        except Exception:
+            return None
+
+    def update_user_role_by_id(self, id: str, role: str) -> Optional[UserModel]:
+        try:
+            with get_db() as db:
+                db.query(User).filter_by(id=id).update({"role": role})
+                db.commit()
+                user = db.query(User).filter_by(id=id).first()
+                return UserModel.model_validate(user)
+        except Exception:
+            return None
+
+    def update_user_profile_image_url_by_id(
+        self, id: str, profile_image_url: str
+    ) -> Optional[UserModel]:
+        try:
+            with get_db() as db:
+                db.query(User).filter_by(id=id).update(
+                    {"profile_image_url": profile_image_url}
+                )
+                db.commit()
+
+                user = db.query(User).filter_by(id=id).first()
+                return UserModel.model_validate(user)
+        except Exception:
+            return None
+
+    def update_user_last_active_by_id(self, id: str) -> Optional[UserModel]:
+        try:
+            with get_db() as db:
+                db.query(User).filter_by(id=id).update(
+                    {"last_active_at": int(time.time())}
+                )
+                db.commit()
+
+                user = db.query(User).filter_by(id=id).first()
+                return UserModel.model_validate(user)
+        except Exception:
+            return None
+
+    def update_user_oauth_sub_by_id(
+        self, id: str, oauth_sub: str
+    ) -> Optional[UserModel]:
+        try:
+            with get_db() as db:
+                db.query(User).filter_by(id=id).update({"oauth_sub": oauth_sub})
+                db.commit()
+
+                user = db.query(User).filter_by(id=id).first()
+                return UserModel.model_validate(user)
+        except Exception:
+            return None
+
+    def update_user_by_id(self, id: str, updated: dict) -> Optional[UserModel]:
+        try:
+            with get_db() as db:
+                db.query(User).filter_by(id=id).update(updated)
+                db.commit()
+
+                user = db.query(User).filter_by(id=id).first()
+                return UserModel.model_validate(user)
+                # return UserModel(**user.dict())
+        except Exception:
+            return None
+
+    def delete_user_by_id(self, id: str) -> bool:
+        try:
+            # Delete User Chats
+            result = Chats.delete_chats_by_user_id(id)
+
+            if result:
+                with get_db() as db:
+                    # Delete User
+                    db.query(User).filter_by(id=id).delete()
+                    db.commit()
+
+                return True
+            else:
+                return False
+        except Exception:
+            return False
+
+    def update_user_api_key_by_id(self, id: str, api_key: str) -> str:
+        try:
+            with get_db() as db:
+                result = db.query(User).filter_by(id=id).update({"api_key": api_key})
+                db.commit()
+                return True if result == 1 else False
+        except Exception:
+            return False
+
+    def get_user_api_key_by_id(self, id: str) -> Optional[str]:
+        try:
+            with get_db() as db:
+                user = db.query(User).filter_by(id=id).first()
+                return user.api_key
+        except Exception:
+            return None
+
+
+Users = UsersTable()

backend/open_webui/apps/webui/routers/auths.py

@@ -1,43 +1,36 @@
-import logging
-
-from fastapi import Request, UploadFile, File
-from fastapi import Depends, HTTPException, status
-from fastapi.responses import Response
-
-from fastapi import APIRouter
-from pydantic import BaseModel
 import re
 import uuid
-import csv
 
-from apps.webui.models.auths import (
-    SigninForm,
-    SignupForm,
+from open_webui.apps.webui.models.auths import (
     AddUserForm,
-    UpdateProfileForm,
-    UpdatePasswordForm,
-    UserResponse,
-    SigninResponse,
-    Auths,
     ApiKey,
+    Auths,
+    SigninForm,
+    SigninResponse,
+    SignupForm,
+    UpdatePasswordForm,
+    UpdateProfileForm,
+    UserResponse,
 )
-from apps.webui.models.users import Users
-
-from utils.utils import (
-    get_password_hash,
-    get_current_user,
-    get_admin_user,
-    create_token,
-    create_api_key,
-)
-from utils.misc import parse_duration, validate_email_format
-from utils.webhook import post_webhook
-from constants import ERROR_MESSAGES, WEBHOOK_MESSAGES
-from config import (
-    WEBUI_AUTH,
+from open_webui.apps.webui.models.users import Users
+from open_webui.config import WEBUI_AUTH
+from open_webui.constants import ERROR_MESSAGES, WEBHOOK_MESSAGES
+from open_webui.env import (
     WEBUI_AUTH_TRUSTED_EMAIL_HEADER,
     WEBUI_AUTH_TRUSTED_NAME_HEADER,
 )
+from fastapi import APIRouter, Depends, HTTPException, Request, status
+from fastapi.responses import Response
+from pydantic import BaseModel
+from open_webui.utils.misc import parse_duration, validate_email_format
+from open_webui.utils.utils import (
+    create_api_key,
+    create_token,
+    get_admin_user,
+    get_current_user,
+    get_password_hash,
+)
+from open_webui.utils.webhook import post_webhook
 
 router = APIRouter()
 
@@ -136,6 +129,7 @@ async def signin(request: Request, response: Response, form_data: SigninForm):
         if not Users.get_user_by_email(trusted_email.lower()):
             await signup(
                 request,
+                response,
                 SignupForm(
                     email=trusted_email, password=str(uuid.uuid4()), name=trusted_name
                 ),
@@ -153,6 +147,7 @@ async def signin(request: Request, response: Response, form_data: SigninForm):
 
             await signup(
                 request,
+                response,
                 SignupForm(email=admin_email, password=admin_password, name="User"),
             )
 
@@ -193,10 +188,19 @@ async def signin(request: Request, response: Response, form_data: SigninForm):
 
 @router.post("/signup", response_model=SigninResponse)
 async def signup(request: Request, response: Response, form_data: SignupForm):
-    if not request.app.state.config.ENABLE_SIGNUP and WEBUI_AUTH:
-        raise HTTPException(
-            status.HTTP_403_FORBIDDEN, detail=ERROR_MESSAGES.ACCESS_PROHIBITED
-        )
+    if WEBUI_AUTH:
+        if (
+            not request.app.state.config.ENABLE_SIGNUP
+            or not request.app.state.config.ENABLE_LOGIN_FORM
+        ):
+            raise HTTPException(
+                status.HTTP_403_FORBIDDEN, detail=ERROR_MESSAGES.ACCESS_PROHIBITED
+            )
+    else:
+        if Users.get_num_users() != 0:
+            raise HTTPException(
+                status.HTTP_403_FORBIDDEN, detail=ERROR_MESSAGES.ACCESS_PROHIBITED
+            )
 
     if not validate_email_format(form_data.email.lower()):
         raise HTTPException(
@@ -226,7 +230,6 @@ async def signup(request: Request, response: Response, form_data: SignupForm):
                 data={"id": user.id},
                 expires_delta=parse_duration(request.app.state.config.JWT_EXPIRES_IN),
             )
-            # response.set_cookie(key='token', value=token, httponly=True)
 
             # Set the cookie token
             response.set_cookie(
@@ -268,7 +271,6 @@ async def signup(request: Request, response: Response, form_data: SignupForm):
 
 @router.post("/add", response_model=SigninResponse)
 async def add_user(form_data: AddUserForm, user=Depends(get_admin_user)):
-
     if not validate_email_format(form_data.email.lower()):
         raise HTTPException(
             status.HTTP_400_BAD_REQUEST, detail=ERROR_MESSAGES.INVALID_EMAIL_FORMAT
@@ -278,7 +280,6 @@ async def add_user(form_data: AddUserForm, user=Depends(get_admin_user)):
         raise HTTPException(400, detail=ERROR_MESSAGES.EMAIL_TAKEN)
 
     try:
-
         print(form_data)
         hashed = get_password_hash(form_data.password)
         user = Auths.insert_new_auth(
@@ -350,6 +351,7 @@ async def get_admin_config(request: Request, user=Depends(get_admin_user)):
         "DEFAULT_USER_ROLE": request.app.state.config.DEFAULT_USER_ROLE,
         "JWT_EXPIRES_IN": request.app.state.config.JWT_EXPIRES_IN,
         "ENABLE_COMMUNITY_SHARING": request.app.state.config.ENABLE_COMMUNITY_SHARING,
+        "ENABLE_MESSAGE_RATING": request.app.state.config.ENABLE_MESSAGE_RATING,
     }
 
 
@@ -359,6 +361,7 @@ class AdminConfig(BaseModel):
     DEFAULT_USER_ROLE: str
     JWT_EXPIRES_IN: str
     ENABLE_COMMUNITY_SHARING: bool
+    ENABLE_MESSAGE_RATING: bool
 
 
 @router.post("/admin/config")
@@ -380,6 +383,7 @@ async def update_admin_config(
     request.app.state.config.ENABLE_COMMUNITY_SHARING = (
         form_data.ENABLE_COMMUNITY_SHARING
     )
+    request.app.state.config.ENABLE_MESSAGE_RATING = form_data.ENABLE_MESSAGE_RATING
 
     return {
         "SHOW_ADMIN_DETAILS": request.app.state.config.SHOW_ADMIN_DETAILS,
@@ -387,6 +391,7 @@ async def update_admin_config(
         "DEFAULT_USER_ROLE": request.app.state.config.DEFAULT_USER_ROLE,
         "JWT_EXPIRES_IN": request.app.state.config.JWT_EXPIRES_IN,
         "ENABLE_COMMUNITY_SHARING": request.app.state.config.ENABLE_COMMUNITY_SHARING,
+        "ENABLE_MESSAGE_RATING": request.app.state.config.ENABLE_MESSAGE_RATING,
     }
 
 

backend/open_webui/apps/webui/routers/chats.py

@@ -1,34 +1,25 @@
-from fastapi import Depends, Request, HTTPException, status
-from datetime import datetime, timedelta
-from typing import List, Union, Optional
-from utils.utils import get_verified_user, get_admin_user
-from fastapi import APIRouter
-from pydantic import BaseModel
 import json
 import logging
+from typing import Optional
 
-from apps.webui.models.users import Users
-from apps.webui.models.chats import (
-    ChatModel,
-    ChatResponse,
-    ChatTitleForm,
+from open_webui.apps.webui.models.chats import (
     ChatForm,
-    ChatTitleIdResponse,
+    ChatResponse,
     Chats,
+    ChatTitleIdResponse,
 )
-
-
-from apps.webui.models.tags import (
-    TagModel,
-    ChatIdTagModel,
+from open_webui.apps.webui.models.tags import (
     ChatIdTagForm,
-    ChatTagsResponse,
+    ChatIdTagModel,
+    TagModel,
     Tags,
 )
-
-from constants import ERROR_MESSAGES
-
-from config import SRC_LOG_LEVELS, ENABLE_ADMIN_EXPORT
+from open_webui.config import ENABLE_ADMIN_CHAT_ACCESS, ENABLE_ADMIN_EXPORT
+from open_webui.constants import ERROR_MESSAGES
+from open_webui.env import SRC_LOG_LEVELS
+from fastapi import APIRouter, Depends, HTTPException, Request, status
+from pydantic import BaseModel
+from open_webui.utils.utils import get_admin_user, get_verified_user
 
 log = logging.getLogger(__name__)
 log.setLevel(SRC_LOG_LEVELS["MODELS"])
@@ -40,12 +31,18 @@ router = APIRouter()
 ############################
 
 
-@router.get("/", response_model=List[ChatTitleIdResponse])
-@router.get("/list", response_model=List[ChatTitleIdResponse])
+@router.get("/", response_model=list[ChatTitleIdResponse])
+@router.get("/list", response_model=list[ChatTitleIdResponse])
 async def get_session_user_chat_list(
-    user=Depends(get_verified_user), skip: int = 0, limit: int = 50
+    user=Depends(get_verified_user), page: Optional[int] = None
 ):
-    return Chats.get_chat_list_by_user_id(user.id, skip, limit)
+    if page is not None:
+        limit = 60
+        skip = (page - 1) * limit
+
+        return Chats.get_chat_title_id_list_by_user_id(user.id, skip=skip, limit=limit)
+    else:
+        return Chats.get_chat_title_id_list_by_user_id(user.id)
 
 
 ############################
@@ -55,7 +52,6 @@ async def get_session_user_chat_list(
 
 @router.delete("/", response_model=bool)
 async def delete_all_user_chats(request: Request, user=Depends(get_verified_user)):
-
     if (
         user.role == "user"
         and not request.app.state.config.USER_PERMISSIONS["chat"]["deletion"]
@@ -74,10 +70,18 @@ async def delete_all_user_chats(request: Request, user=Depends(get_verified_user
 ############################
 
 
-@router.get("/list/user/{user_id}", response_model=List[ChatTitleIdResponse])
+@router.get("/list/user/{user_id}", response_model=list[ChatTitleIdResponse])
 async def get_user_chat_list_by_user_id(
-    user_id: str, user=Depends(get_admin_user), skip: int = 0, limit: int = 50
+    user_id: str,
+    user=Depends(get_admin_user),
+    skip: int = 0,
+    limit: int = 50,
 ):
+    if not ENABLE_ADMIN_CHAT_ACCESS:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail=ERROR_MESSAGES.ACCESS_PROHIBITED,
+        )
     return Chats.get_chat_list_by_user_id(
         user_id, include_archived=True, skip=skip, limit=limit
     )
@@ -105,7 +109,7 @@ async def create_new_chat(form_data: ChatForm, user=Depends(get_verified_user)):
 ############################
 
 
-@router.get("/all", response_model=List[ChatResponse])
+@router.get("/all", response_model=list[ChatResponse])
 async def get_user_chats(user=Depends(get_verified_user)):
     return [
         ChatResponse(**{**chat.model_dump(), "chat": json.loads(chat.chat)})
@@ -118,8 +122,8 @@ async def get_user_chats(user=Depends(get_verified_user)):
 ############################
 
 
-@router.get("/all/archived", response_model=List[ChatResponse])
-async def get_user_chats(user=Depends(get_verified_user)):
+@router.get("/all/archived", response_model=list[ChatResponse])
+async def get_user_archived_chats(user=Depends(get_verified_user)):
     return [
         ChatResponse(**{**chat.model_dump(), "chat": json.loads(chat.chat)})
         for chat in Chats.get_archived_chats_by_user_id(user.id)
@@ -131,7 +135,7 @@ async def get_user_chats(user=Depends(get_verified_user)):
 ############################
 
 
-@router.get("/all/db", response_model=List[ChatResponse])
+@router.get("/all/db", response_model=list[ChatResponse])
 async def get_all_user_chats_in_db(user=Depends(get_admin_user)):
     if not ENABLE_ADMIN_EXPORT:
         raise HTTPException(
@@ -149,7 +153,7 @@ async def get_all_user_chats_in_db(user=Depends(get_admin_user)):
 ############################
 
 
-@router.get("/archived", response_model=List[ChatTitleIdResponse])
+@router.get("/archived", response_model=list[ChatTitleIdResponse])
 async def get_archived_session_user_chat_list(
     user=Depends(get_verified_user), skip: int = 0, limit: int = 50
 ):
@@ -178,9 +182,9 @@ async def get_shared_chat_by_id(share_id: str, user=Depends(get_verified_user)):
             status_code=status.HTTP_401_UNAUTHORIZED, detail=ERROR_MESSAGES.NOT_FOUND
         )
 
-    if user.role == "user":
+    if user.role == "user" or (user.role == "admin" and not ENABLE_ADMIN_CHAT_ACCESS):
         chat = Chats.get_chat_by_share_id(share_id)
-    elif user.role == "admin":
+    elif user.role == "admin" and ENABLE_ADMIN_CHAT_ACCESS:
         chat = Chats.get_chat_by_id(share_id)
 
     if chat:
@@ -202,12 +206,10 @@ class TagNameForm(BaseModel):
     limit: Optional[int] = 50
 
 
-@router.post("/tags", response_model=List[ChatTitleIdResponse])
+@router.post("/tags", response_model=list[ChatTitleIdResponse])
 async def get_user_chat_list_by_tag_name(
     form_data: TagNameForm, user=Depends(get_verified_user)
 ):
-
-    print(form_data)
     chat_ids = [
         chat_id_tag.chat_id
         for chat_id_tag in Tags.get_chat_ids_by_tag_name_and_user_id(
@@ -228,7 +230,7 @@ async def get_user_chat_list_by_tag_name(
 ############################
 
 
-@router.get("/tags/all", response_model=List[TagModel])
+@router.get("/tags/all", response_model=list[TagModel])
 async def get_all_tags(user=Depends(get_verified_user)):
     try:
         tags = Tags.get_tags_by_user_id(user.id)
@@ -286,7 +288,6 @@ async def update_chat_by_id(
 
 @router.delete("/{id}", response_model=bool)
 async def delete_chat_by_id(request: Request, id: str, user=Depends(get_verified_user)):
-
     if user.role == "admin":
         result = Chats.delete_chat_by_id(id)
         return result
@@ -310,7 +311,6 @@ async def delete_chat_by_id(request: Request, id: str, user=Depends(get_verified
 async def clone_chat_by_id(id: str, user=Depends(get_verified_user)):
     chat = Chats.get_chat_by_id_and_user_id(id, user.id)
     if chat:
-
         chat_body = json.loads(chat.chat)
         updated_chat = {
             **chat_body,
@@ -404,7 +404,7 @@ async def delete_shared_chat_by_id(id: str, user=Depends(get_verified_user)):
 ############################
 
 
-@router.get("/{id}/tags", response_model=List[TagModel])
+@router.get("/{id}/tags", response_model=list[TagModel])
 async def get_chat_tags_by_id(id: str, user=Depends(get_verified_user)):
     tags = Tags.get_tags_by_chat_id_and_user_id(id, user.id)