[PR #24031] [CLOSED] perf: Remake RAG citations to enable prompt caching #114797

Closed
opened 2026-05-18 15:38:07 -05:00 by GiteaMirror · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/open-webui/open-webui/pull/24031
Author: @Lumpiasty
Created: 4/23/2026
Status: Closed

Base: devHead: dev


📝 Commits (3)

  • 65b8c07 feat(rag): refactor agentic RAG context injection for cache stability
  • fe0cfaa feat(rag): persist non-agentic RAG context across turns for caching
  • 1d582ac Drop unused rag_template function

📊 Changes

8 files changed (+201 additions, -133 deletions)

View changed files

📝 backend/open_webui/config.py (+6 -9)
📝 backend/open_webui/env.py (+0 -2)
📝 backend/open_webui/main.py (+3 -3)
📝 backend/open_webui/routers/retrieval.py (+6 -6)
📝 backend/open_webui/utils/middleware.py (+167 -69)
📝 backend/open_webui/utils/misc.py (+18 -0)
📝 backend/open_webui/utils/task.py (+0 -43)
📝 src/lib/components/admin/Settings/Documents.svelte (+1 -1)

📄 Description

Pull Request Checklist

Note to first-time contributors: Please open a discussion post in Discussions to discuss your idea/fix with the community before creating a pull request, and describe your changes before submitting a pull request.

This is to ensure large feature PRs are discussed with the community first, before starting work on it. If the community does not want this feature or it is not relevant for Open WebUI as a project, it can be identified in the discussion before working on the feature and submitting the PR.

Before submitting, make sure you've checked the following:

  • Target branch: Verify that the pull request targets the dev branch. PRs targeting main will be immediately closed.
  • Description: Provide a concise description of the changes made in this pull request down below.
  • Changelog: Ensure a changelog entry following the format of Keep a Changelog is added at the bottom of the PR description.
  • Documentation: Add docs in Open WebUI Docs Repository. Document user-facing behavior, environment variables, public APIs/interfaces, or deployment steps.
  • Dependencies: Are there any new or upgraded dependencies? If so, explain why, update the changelog/docs, and include any compatibility notes. Actually run the code/function that uses updated library to ensure it doesn't crash.
  • Testing: Perform manual tests to verify the implemented fix/feature works as intended AND does not break any other functionality. Include reproducible steps to demonstrate the issue before the fix. Test edge cases (URL encoding, HTML entities, types). Take this as an opportunity to make screenshots of the feature/fix and include them in the PR description.
  • Agentic AI Code: Confirm this Pull Request is not written by any AI Agent or has at least gone through additional human review AND manual testing. If any AI Agent is the co-author of this PR, it may lead to immediate closure of the PR.
  • Code review: Have you performed a self-review of your code, addressing any coding standard issues and ensuring adherence to the project's coding standards?
  • Design & Architecture: Prefer smart defaults over adding new settings; use local state for ephemeral UI logic. Open a Discussion for major architectural or UX changes.
  • Git Hygiene: Keep PRs atomic (one logical change). Clean up commits and rebase on dev to ensure no unrelated commits (e.g. from main) are included. Push updates to the existing PR branch instead of closing and reopening.
  • Title Prefix: To clearly categorize this pull request, prefix the pull request title using one of the following:
    • BREAKING CHANGE: Significant changes that may affect compatibility
    • build: Changes that affect the build system or external dependencies
    • ci: Changes to our continuous integration processes or workflows
    • chore: Refactor, cleanup, or other non-functional code changes
    • docs: Documentation update or addition
    • feat: Introduces a new feature or enhancement to the codebase
    • fix: Bug fix or error correction
    • i18n: Internationalization or localization changes
    • perf: Performance improvement
    • refactor: Code restructuring for better maintainability, readability, or scalability
    • style: Changes that do not affect the meaning of the code (white space, formatting, missing semi-colons, etc.)
    • test: Adding missing tests or correcting existing tests
    • WIP: Work in progress, a temporary label for incomplete or ongoing work

Changelog Entry

Description

Current RAG citation system system modifies user message, which invalidates prompt cache which in turn explodes your inference bills and/or makes you wait for prompt processing way longer than necessary, especially noticable when fetching multiple large documents.
I've redesigned how citations work (injecting instructions, source references) to be append-only, enabling caching.

Added

none

Changed

  • Context source references and RAG instructions now are persisted to database using two new hidden message types:
    • open_webui:rag_source
      • gets serialized like tool call output
      • contains rag instructions on first instance
      • contains context source references <source id='x'> of last tool output in every instance
      • used in agentic code path after retrieval tools output
    • open_webui:rag_context
      • gets serialized as <context> prepending following user message
      • contains only source references and snippets
      • if present in chat history, rag instructions get inserted in system message
      • used in non-agentic path

Deprecated

none

Removed

none

Fixed

  • App no longer modifies user message or system prompt on the fly when doing agentic research enabling context caching
  • name parameter of the <source> tag now contains name of the specified source instead of tool name improving citation accuracy

Security

none

Breaking Changes

  • BREAKING CHANGE: Removed RAG_SYSTEM_CONTEXT configuration option as now redundant
  • BREAKING CHANGE: Changed RAG_TEMPLATE to match new system, removed {{CONTEXT}} variable and renamed option to RAG_INSTRUCTIONS (rag.instructions in db) to avoid using old template if user changed it previously

Additional Information

  • When testing I noticed that even though this wasn't the main motivation, this change even improves quality of responses, especially the non-agentic path, small local models like Qwen3.5-4B previously spent significant amount of time wondering what to do with my weird request ("Respond to user message..."), now they just acknowledge context and get to the point.

Screenshots or Videos

No changes in the UI, only fixing the way backend uses inference api.

Contributor License Agreement

Note

Deleting the CLA section will lead to immediate closure of your PR and it will not be merged in.


🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/open-webui/open-webui/pull/24031 **Author:** [@Lumpiasty](https://github.com/Lumpiasty) **Created:** 4/23/2026 **Status:** ❌ Closed **Base:** `dev` ← **Head:** `dev` --- ### 📝 Commits (3) - [`65b8c07`](https://github.com/open-webui/open-webui/commit/65b8c0753d0dfdddbc5f1715fb75926c96f0fca8) feat(rag): refactor agentic RAG context injection for cache stability - [`fe0cfaa`](https://github.com/open-webui/open-webui/commit/fe0cfaaa79a680bd4b7b2d0f85c09a3ef13043b4) feat(rag): persist non-agentic RAG context across turns for caching - [`1d582ac`](https://github.com/open-webui/open-webui/commit/1d582ac9c274cab5e7c87f931af726330b5a5b42) Drop unused rag_template function ### 📊 Changes **8 files changed** (+201 additions, -133 deletions) <details> <summary>View changed files</summary> 📝 `backend/open_webui/config.py` (+6 -9) 📝 `backend/open_webui/env.py` (+0 -2) 📝 `backend/open_webui/main.py` (+3 -3) 📝 `backend/open_webui/routers/retrieval.py` (+6 -6) 📝 `backend/open_webui/utils/middleware.py` (+167 -69) 📝 `backend/open_webui/utils/misc.py` (+18 -0) 📝 `backend/open_webui/utils/task.py` (+0 -43) 📝 `src/lib/components/admin/Settings/Documents.svelte` (+1 -1) </details> ### 📄 Description <!-- ⚠️ CRITICAL CHECKS FOR CONTRIBUTORS (READ, DON'T DELETE) ⚠️ 1. Target the `dev` branch. PRs targeting `main` will be automatically closed. 2. Do NOT delete the CLA section at the bottom. It is required for the bot to accept your PR. --> # Pull Request Checklist ### Note to first-time contributors: Please open a discussion post in [Discussions](https://github.com/open-webui/open-webui/discussions) to discuss your idea/fix with the community before creating a pull request, and describe your changes before submitting a pull request. This is to ensure large feature PRs are discussed with the community first, before starting work on it. If the community does not want this feature or it is not relevant for Open WebUI as a project, it can be identified in the discussion before working on the feature and submitting the PR. <!-- ### ⚠️ Important: Your PR is a contribution, not a guarantee of merge. The most impactful way to contribute to Open WebUI is through well-written bug reports, detailed feature discussions, and thoughtful ideas. These directly shape the project. If you do open a pull request, please know that Open WebUI is held to the highest standard of code quality, consistency, and architectural coherence, and every line merged becomes something the core team must own, maintain, and support indefinitely. Submitted code may be refactored, rewritten, or used as inspiration for a different implementation. This is not a reflection of your work's quality. It is how we ensure that a small team can deeply understand and evolve every part of the codebase. --> **Before submitting, make sure you've checked the following:** - [x] **Target branch:** Verify that the pull request targets the `dev` branch. **PRs targeting `main` will be immediately closed.** - [x] **Description:** Provide a concise description of the changes made in this pull request down below. - [x] **Changelog:** Ensure a changelog entry following the format of [Keep a Changelog](https://keepachangelog.com/) is added at the bottom of the PR description. - [ ] **Documentation:** Add docs in [Open WebUI Docs Repository](https://github.com/open-webui/docs). Document user-facing behavior, environment variables, public APIs/interfaces, or deployment steps. - [x] **Dependencies:** Are there any new or upgraded dependencies? If so, explain why, update the changelog/docs, and include any compatibility notes. Actually run the code/function that uses updated library to ensure it doesn't crash. - [x] **Testing:** Perform manual tests to **verify the implemented fix/feature works as intended AND does not break any other functionality**. Include reproducible steps to demonstrate the issue before the fix. Test edge cases (URL encoding, HTML entities, types). Take this as an opportunity to **make screenshots of the feature/fix and include them in the PR description**. - [x] **Agentic AI Code:** Confirm this Pull Request is **not written by any AI Agent** or has at least **gone through additional human review AND manual testing**. If any AI Agent is the co-author of this PR, it may lead to immediate closure of the PR. - [x] **Code review:** Have you performed a self-review of your code, addressing any coding standard issues and ensuring adherence to the project's coding standards? - [x] **Design & Architecture:** Prefer smart defaults over adding new settings; use local state for ephemeral UI logic. Open a Discussion for major architectural or UX changes. - [x] **Git Hygiene:** Keep PRs atomic (one logical change). Clean up commits and rebase on `dev` to ensure no unrelated commits (e.g. from `main`) are included. Push updates to the existing PR branch instead of closing and reopening. - [x] **Title Prefix:** To clearly categorize this pull request, prefix the pull request title using one of the following: - **BREAKING CHANGE**: Significant changes that may affect compatibility - **build**: Changes that affect the build system or external dependencies - **ci**: Changes to our continuous integration processes or workflows - **chore**: Refactor, cleanup, or other non-functional code changes - **docs**: Documentation update or addition - **feat**: Introduces a new feature or enhancement to the codebase - **fix**: Bug fix or error correction - **i18n**: Internationalization or localization changes - **perf**: Performance improvement - **refactor**: Code restructuring for better maintainability, readability, or scalability - **style**: Changes that do not affect the meaning of the code (white space, formatting, missing semi-colons, etc.) - **test**: Adding missing tests or correcting existing tests - **WIP**: Work in progress, a temporary label for incomplete or ongoing work # Changelog Entry ### Description Current RAG citation system system modifies user message, which invalidates prompt cache which in turn explodes your inference bills and/or makes you wait for prompt processing way longer than necessary, especially noticable when fetching multiple large documents. I've redesigned how citations work (injecting instructions, source references) to be append-only, enabling caching. ### Added none ### Changed - Context source references and RAG instructions now are persisted to database using two new hidden message types: - `open_webui:rag_source` - gets serialized like tool call output - contains rag instructions on first instance - contains context source references `<source id='x'>` of last tool output in every instance - used in agentic code path after retrieval tools output - `open_webui:rag_context` - gets serialized as `<context>` prepending following user message - contains only source references and snippets - if present in chat history, rag instructions get inserted in system message - used in non-agentic path ### Deprecated none ### Removed none ### Fixed - App no longer modifies user message or system prompt on the fly when doing agentic research enabling context caching - `name` parameter of the `<source>` tag now contains name of the specified source instead of tool name improving citation accuracy ### Security none ### Breaking Changes - **BREAKING CHANGE**: Removed `RAG_SYSTEM_CONTEXT` configuration option as now redundant - **BREAKING CHANGE**: Changed `RAG_TEMPLATE` to match new system, removed `{{CONTEXT}}` variable and renamed option to `RAG_INSTRUCTIONS` (`rag.instructions` in db) to avoid using old template if user changed it previously --- ### Additional Information - When testing I noticed that even though this wasn't the main motivation, this change even improves quality of responses, especially the non-agentic path, small local models like Qwen3.5-4B previously spent significant amount of time wondering what to do with my weird request ("Respond to user message..."), now they just acknowledge context and get to the point. ### Screenshots or Videos No changes in the UI, only fixing the way backend uses inference api. ### Contributor License Agreement <!-- 🚨 DO NOT DELETE THE TEXT BELOW 🚨 Keep the "Contributor License Agreement" confirmation text intact. Deleting it will trigger the CLA-Bot to INVALIDATE your PR. Your PR will NOT be reviewed or merged until you check the box below confirming that you have read and agree to the terms of the CLA. --> - [x] By submitting this pull request, I confirm that I have read and fully agree to the [Contributor License Agreement (CLA)](https://github.com/open-webui/open-webui/blob/main/CONTRIBUTOR_LICENSE_AGREEMENT), and I am providing my contributions under its terms. > [!NOTE] > Deleting the CLA section will lead to immediate closure of your PR and it will not be merged in. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
GiteaMirror added the pull-request label 2026-05-18 15:38:07 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#114797