github-starred/open-webui
2
0
Fork 0
mirror of https://github.com/open-webui/open-webui.git synced 2026-07-15 21:19:39 -05:00
Code Issues 2.4k Packages Projects Releases 129 Wiki Activity

[GH-ISSUE #14543] issue: Custom models based on those from OpenAI compatible API are not visible in the models list #17298

New Issue
Closed
opened 2026-04-19 23:01:29 -05:00 by GiteaMirror · 10 comments
GiteaMirror commented 2026-04-19 23:01:29 -05:00
Owner
Copy Link

Originally created by @cybergod96 on GitHub (May 31, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/14543

Check Existing Issues

  • I have searched the existing issues and discussions.
  • I am using the latest version of Open WebUI.

Installation Method

Pip Install

Open WebUI Version

0.6.13

Ollama Version (if applicable)

No response

Operating System

Windows 10

Browser (if applicable)

Firefox 139.0.1

Confirmation

  • I have read and followed all instructions in README.md.
  • I am using 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 every relevant configuration, setting, and environment variable used in my setup.
  • I have clearly listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc).
  • I have documented step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation. My steps:
  • Start with the initial platform/version/OS and dependencies used,
  • Specify exact install/launch/configure commands,
  • List URLs visited, user input (incl. example values/emails/passwords if needed),
  • Describe all options and toggles enabled or changed,
  • Include any files or environmental changes,
  • Identify the expected and actual result at each stage,
  • Ensure any reasonably skilled user can follow and hit the same issue.

Expected Behavior

Custom model (created in Workspace > Models) based on model from OpenAI compatible API should be visible on the list of models in the Chat window.

Actual Behavior

Custom model (created in Workspace > Models) based on model from OpenAI compatible API is not visible on the list of models in the Chat window.

Steps to Reproduce

  1. Go to Settings > Connections
  2. Add connection to OpenAI compatible API and insert some valid model ids (eg. deepseek-ai/DeepSeek-R1-0528)
  3. Go to Workspace > Models
  4. Click + icon
  5. In the Base model select deepseek-ai/DeepSeek-R1-0528
  6. Set name to R1 Custom (r1-custom)
  7. Set model visibility to Public
  8. Edit other model settings like changing Temperature to value other than default
  9. Click Save & Update. The newly created model is visible on the list of models in the Workspace > Models window
  10. Click New Chat
  11. Open list of models. The r1-custom model is not visible on the list.

Logs & Screenshots

Image

Image

Image

Image

Backend logs:

2025-05-31 10:50:50.587 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53985 - "GET /api/models HTTP/1.1" 200 - {}
2025-05-31 10:50:50.601 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53979 - "GET /api/v1/configs/banners HTTP/1.1" 200 - {}
2025-05-31 10:50:50.614 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53979 - "GET /api/v1/tools/ HTTP/1.1" 200 - {}
2025-05-31 10:50:50.685 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53985 - "GET /api/v1/users/user/settings HTTP/1.1" 200 - {}
2025-05-31 10:50:50.747 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53985 - "POST /api/v1/folders/0a6fd46c-cad4-48df-b22e-5aadb591d27f/update/expanded HTTP/1.1" 200 - {}
2025-05-31 10:50:50.822 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53983 - "POST /api/v1/folders/6579aba1-f420-4ad6-aa26-67795d7034f4/update/expanded HTTP/1.1" 200 - {}
2025-05-31 10:50:50.823 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53984 - "POST /api/v1/folders/db58110a-19f0-4c29-bfa9-f4488b3da976/update/expanded HTTP/1.1" 200 - {}
2025-05-31 10:50:50.824 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53981 - "POST /api/v1/folders/43d6840a-3f2a-4f9b-b849-8cbbf75442f9/update/expanded HTTP/1.1" 200 - {}
2025-05-31 10:50:50.829 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53992 - "POST /api/v1/folders/23eb1732-43a2-41cd-a715-066f6f355979/update/expanded HTTP/1.1" 200 - {}
2025-05-31 10:50:50.844 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53985 - "POST /api/v1/folders/024cfeaa-035b-4ba0-aabc-d912e287c3a7/update/expanded HTTP/1.1" 200 - {}
2025-05-31 10:50:50.883 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53979 - "GET /ollama/api/version HTTP/1.1" 200 - {}
2025-05-31 10:51:06.011 | INFO     | open_webui.routers.ollama:get_all_models:323 - get_all_models() - {}
2025-05-31 10:51:06.531 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53995 - "GET /api/models HTTP/1.1" 200 - {}
2025-05-31 10:51:17.682 | INFO     | open_webui.routers.ollama:get_all_models:323 - get_all_models() - {}
2025-05-31 10:51:18.231 | INFO     | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:54000 - "GET /api/models HTTP/1.1" 200 - {}

Browser logs:

0:53:03.489 Socket undefined disconnected due to transport close [0.BVEZraL0.js:1:47470](http://localhost:8080/_app/immutable/nodes/0.BVEZraL0.js)
10:53:03.490 Additional details: 
Object { description: "websocket connection closed", context: close }
[0.BVEZraL0.js:1:47528](http://localhost:8080/_app/immutable/nodes/0.BVEZraL0.js)
10:53:04.451 Backend config: 
Object { status: true, name: "Open WebUI", version: "0.6.13", default_locale: "", oauth: {…}, features: {…}, default_models: null, default_prompt_suggestions: [], user_count: 1, code: {…}, … }
[+layout.svelte:559:11](http://localhost:8080/src/routes/+layout.svelte)
10:53:04.505 connected zMXhZfPSLLw-EmDOAAAN [+layout.svelte:89:11](http://localhost:8080/src/routes/+layout.svelte)
10:53:04.505 user-list 
Object { user_ids: (1) […] }
[+layout.svelte:108:11](http://localhost:8080/src/routes/+layout.svelte)
10:53:04.506 usage 
Object { models: [] }
[+layout.svelte:113:11](http://localhost:8080/src/routes/+layout.svelte)
10:53:04.525 user-list 
Object { user_ids: (1) […] }
[+layout.svelte:108:11](http://localhost:8080/src/routes/+layout.svelte)
10:53:05.262 mounted [Chat.svelte:429:10](http://localhost:8080/src/lib/components/chat/Chat.svelte)
10:53:05.304 content <empty string> [RichTextInput.svelte:192:10](http://localhost:8080/src/lib/components/common/RichTextInput.svelte)
10:53:05.305 [tiptap warn]: Duplicate extension names found: ['codeBlock']. This can lead to issues. [index.js:1183:20](http://localhost:8080/node_modules/@tiptap/core/dist/index.js)

Additional Information

  • Ollama was initially disabled in Admin settings but enabling it does not fix the issue
  • The connection to the OpenAI compatible API itself works fine
  • I have tried to restart app and refresh browser window but this does not fix the issue
  • I have checked the documentation and it does not state that custom models are for ollama only so I assumed it should work
Originally created by @cybergod96 on GitHub (May 31, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/14543 ### Check Existing Issues - [x] I have searched the existing issues and discussions. - [x] I am using the latest version of Open WebUI. ### Installation Method Pip Install ### Open WebUI Version 0.6.13 ### Ollama Version (if applicable) _No response_ ### Operating System Windows 10 ### Browser (if applicable) Firefox 139.0.1 ### Confirmation - [x] I have read and followed all instructions in `README.md`. - [x] I am using the latest version of **both** Open WebUI and Ollama. - [x] I have included the browser console logs. - [x] I have included the Docker container logs. - [x] I have **provided every relevant configuration, setting, and environment variable used in my setup.** - [x] I have clearly **listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup** (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc). - [x] I have documented **step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation**. My steps: - Start with the initial platform/version/OS and dependencies used, - Specify exact install/launch/configure commands, - List URLs visited, user input (incl. example values/emails/passwords if needed), - Describe all options and toggles enabled or changed, - Include any files or environmental changes, - Identify the expected and actual result at each stage, - Ensure any reasonably skilled user can follow and hit the same issue. ### Expected Behavior Custom model (created in Workspace > Models) based on model from OpenAI compatible API should be visible on the list of models in the Chat window. ### Actual Behavior Custom model (created in Workspace > Models) based on model from OpenAI compatible API is not visible on the list of models in the Chat window. ### Steps to Reproduce 1. Go to Settings > Connections 2. Add connection to OpenAI compatible API and insert some valid model ids (eg. deepseek-ai/DeepSeek-R1-0528) 3. Go to Workspace > Models 4. Click + icon 5. In the Base model select deepseek-ai/DeepSeek-R1-0528 6. Set name to R1 Custom (r1-custom) 7. Set model visibility to Public 8. Edit other model settings like changing Temperature to value other than default 9. Click Save & Update. The newly created model is visible on the list of models in the Workspace > Models window 10. Click New Chat 11. Open list of models. The r1-custom model is not visible on the list. ### Logs & Screenshots ![Image](https://github.com/user-attachments/assets/1c0291f9-147a-4d49-b5ac-b3137b6efed5) ![Image](https://github.com/user-attachments/assets/5fb9e274-f15b-4a7b-a838-4dc1fe0c28d9) ![Image](https://github.com/user-attachments/assets/eeb853ce-ff69-4a93-9971-bc9efdcf1d64) ![Image](https://github.com/user-attachments/assets/b23f7f0d-16ff-42bd-b029-f9c4de15ab9e) Backend logs: ``` 2025-05-31 10:50:50.587 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53985 - "GET /api/models HTTP/1.1" 200 - {} 2025-05-31 10:50:50.601 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53979 - "GET /api/v1/configs/banners HTTP/1.1" 200 - {} 2025-05-31 10:50:50.614 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53979 - "GET /api/v1/tools/ HTTP/1.1" 200 - {} 2025-05-31 10:50:50.685 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53985 - "GET /api/v1/users/user/settings HTTP/1.1" 200 - {} 2025-05-31 10:50:50.747 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53985 - "POST /api/v1/folders/0a6fd46c-cad4-48df-b22e-5aadb591d27f/update/expanded HTTP/1.1" 200 - {} 2025-05-31 10:50:50.822 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53983 - "POST /api/v1/folders/6579aba1-f420-4ad6-aa26-67795d7034f4/update/expanded HTTP/1.1" 200 - {} 2025-05-31 10:50:50.823 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53984 - "POST /api/v1/folders/db58110a-19f0-4c29-bfa9-f4488b3da976/update/expanded HTTP/1.1" 200 - {} 2025-05-31 10:50:50.824 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53981 - "POST /api/v1/folders/43d6840a-3f2a-4f9b-b849-8cbbf75442f9/update/expanded HTTP/1.1" 200 - {} 2025-05-31 10:50:50.829 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53992 - "POST /api/v1/folders/23eb1732-43a2-41cd-a715-066f6f355979/update/expanded HTTP/1.1" 200 - {} 2025-05-31 10:50:50.844 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53985 - "POST /api/v1/folders/024cfeaa-035b-4ba0-aabc-d912e287c3a7/update/expanded HTTP/1.1" 200 - {} 2025-05-31 10:50:50.883 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53979 - "GET /ollama/api/version HTTP/1.1" 200 - {} 2025-05-31 10:51:06.011 | INFO | open_webui.routers.ollama:get_all_models:323 - get_all_models() - {} 2025-05-31 10:51:06.531 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:53995 - "GET /api/models HTTP/1.1" 200 - {} 2025-05-31 10:51:17.682 | INFO | open_webui.routers.ollama:get_all_models:323 - get_all_models() - {} 2025-05-31 10:51:18.231 | INFO | uvicorn.protocols.http.httptools_impl:send:476 - 127.0.0.1:54000 - "GET /api/models HTTP/1.1" 200 - {} ``` Browser logs: ``` 0:53:03.489 Socket undefined disconnected due to transport close [0.BVEZraL0.js:1:47470](http://localhost:8080/_app/immutable/nodes/0.BVEZraL0.js) 10:53:03.490 Additional details: Object { description: "websocket connection closed", context: close } [0.BVEZraL0.js:1:47528](http://localhost:8080/_app/immutable/nodes/0.BVEZraL0.js) 10:53:04.451 Backend config: Object { status: true, name: "Open WebUI", version: "0.6.13", default_locale: "", oauth: {…}, features: {…}, default_models: null, default_prompt_suggestions: [], user_count: 1, code: {…}, … } [+layout.svelte:559:11](http://localhost:8080/src/routes/+layout.svelte) 10:53:04.505 connected zMXhZfPSLLw-EmDOAAAN [+layout.svelte:89:11](http://localhost:8080/src/routes/+layout.svelte) 10:53:04.505 user-list Object { user_ids: (1) […] } [+layout.svelte:108:11](http://localhost:8080/src/routes/+layout.svelte) 10:53:04.506 usage Object { models: [] } [+layout.svelte:113:11](http://localhost:8080/src/routes/+layout.svelte) 10:53:04.525 user-list Object { user_ids: (1) […] } [+layout.svelte:108:11](http://localhost:8080/src/routes/+layout.svelte) 10:53:05.262 mounted [Chat.svelte:429:10](http://localhost:8080/src/lib/components/chat/Chat.svelte) 10:53:05.304 content <empty string> [RichTextInput.svelte:192:10](http://localhost:8080/src/lib/components/common/RichTextInput.svelte) 10:53:05.305 [tiptap warn]: Duplicate extension names found: ['codeBlock']. This can lead to issues. [index.js:1183:20](http://localhost:8080/node_modules/@tiptap/core/dist/index.js) ``` ### Additional Information - Ollama was initially disabled in Admin settings but enabling it does not fix the issue - The connection to the OpenAI compatible API itself works fine - I have tried to restart app and refresh browser window but this does not fix the issue - I have checked the documentation and it does not state that custom models are for ollama only so I assumed it should work
GiteaMirror added the bug label 2026-04-19 23:01:29 -05:00
GiteaMirror commented 2026-04-19 23:01:34 -05:00
Author
Owner
Copy Link

@silenceroom commented on GitHub (Jun 1, 2025):

The new updates now supports to "hide" the custom workspace model from the "dropdown" model list, see if those models were hided from the workspace operation.

<!-- gh-comment-id:2926583898 --> @silenceroom commented on GitHub (Jun 1, 2025): The new updates now supports to "hide" the custom workspace model from the "dropdown" model list, see if those models were hided from the workspace operation.
GiteaMirror commented 2026-04-19 23:01:36 -05:00
Author
Owner
Copy Link

@firmansi commented on GitHub (Jun 1, 2025):

I don't experience this so far, even with quite buggy latest version, 0.6.12 and 0.6.13. I use openrouter as the API

<!-- gh-comment-id:2926714467 --> @firmansi commented on GitHub (Jun 1, 2025): I don't experience this so far, even with quite buggy latest version, 0.6.12 and 0.6.13. I use openrouter as the API
GiteaMirror commented 2026-04-19 23:01:39 -05:00
Author
Owner
Copy Link

@cybergod96 commented on GitHub (Jun 1, 2025):

This didn't help unfortunately. I disabled and enabled model, then clicked Hide and then Show. I even tried exporting it, deleting and importing, but still the model is not visible on the list.

Here are debug logs:
logs_no_ollama.txt - ollama was disabled
logs_with_ollama.txt - ollama was enabled

Endpoint /api/models returns empty list.

<!-- gh-comment-id:2926718890 --> @cybergod96 commented on GitHub (Jun 1, 2025): This didn't help unfortunately. I disabled and enabled model, then clicked Hide and then Show. I even tried exporting it, deleting and importing, but still the model is not visible on the list. Here are debug logs: [logs_no_ollama.txt](https://github.com/user-attachments/files/20538680/logs_no_ollama.txt) - ollama was disabled [logs_with_ollama.txt](https://github.com/user-attachments/files/20538679/logs_with_ollama.txt) - ollama was enabled Endpoint /api/models returns empty list.
GiteaMirror commented 2026-04-19 23:01:41 -05:00
Author
Owner
Copy Link

@firmansi commented on GitHub (Jun 1, 2025):

Are you using public IP or private IP?

<!-- gh-comment-id:2926722855 --> @firmansi commented on GitHub (Jun 1, 2025): Are you using public IP or private IP?
GiteaMirror commented 2026-04-19 23:01:41 -05:00
Author
Owner
Copy Link

@cybergod96 commented on GitHub (Jun 1, 2025):

Private IP.

<!-- gh-comment-id:2926725497 --> @cybergod96 commented on GitHub (Jun 1, 2025): Private IP.
GiteaMirror commented 2026-04-19 23:01:42 -05:00
Author
Owner
Copy Link

@firmansi commented on GitHub (Jun 1, 2025):

In this case, my openwebui running with public IP and using the from Openrouter and don't have any issue with Custom Model with base model Deepseek R1-0528. I know this is not a solution as you might expect, but you might want to use custom pipe. OpenwebUI sometimes experience issues from one version to another latest version since so many developers contribute their work to openWebUI, so sometimes new features addition can cause unexpected issues that not happening before

<!-- gh-comment-id:2926757150 --> @firmansi commented on GitHub (Jun 1, 2025): In this case, my openwebui running with public IP and using the from Openrouter and don't have any issue with Custom Model with base model Deepseek R1-0528. I know this is not a solution as you might expect, but you might want to use custom pipe. OpenwebUI sometimes experience issues from one version to another latest version since so many developers contribute their work to openWebUI, so sometimes new features addition can cause unexpected issues that not happening before
GiteaMirror commented 2026-04-19 23:01:44 -05:00
Author
Owner
Copy Link

@cybergod96 commented on GitHub (Jun 1, 2025):

Okay, after a bit of debugging I found that function get_all_models in backend/open_webui/utils/models.py prematurely returns empty list if function get_all_base_models returns empty list and it does when ollama and openai api integration is disabled (which is my case). After commenting out this return, the model shows up on the list but then when I try to chat with it I get "Model not found" error. Gonna do some more debugging and maybe I'll be able to fix the issue.

<!-- gh-comment-id:2927493335 --> @cybergod96 commented on GitHub (Jun 1, 2025): Okay, after a bit of debugging I found that function get_all_models in backend/open_webui/utils/models.py prematurely returns empty list if function get_all_base_models returns empty list and it does when ollama and openai api integration is disabled (which is my case). After commenting out this return, the model shows up on the list but then when I try to chat with it I get "Model not found" error. Gonna do some more debugging and maybe I'll be able to fix the issue.
GiteaMirror commented 2026-04-19 23:01:45 -05:00
Author
Owner
Copy Link

@Yorickbao commented on GitHub (Jun 3, 2025):

Yes, I have also encountered this problem. The model disappeared inexplicably. Upgrade to 0.6.13 today

<!-- gh-comment-id:2934286502 --> @Yorickbao commented on GitHub (Jun 3, 2025): Yes, I have also encountered this problem. The model disappeared inexplicably. Upgrade to 0.6.13 today
GiteaMirror commented 2026-04-19 23:01:46 -05:00
Author
Owner
Copy Link

@Zyfax commented on GitHub (Jun 3, 2025):

You need to add the connection under Admin Panel > Settings > Connection not under Settings > Connections.

<!-- gh-comment-id:2934377621 --> @Zyfax commented on GitHub (Jun 3, 2025): You need to add the connection under `Admin Panel > Settings > Connection` not under `Settings > Connections`.
GiteaMirror commented 2026-04-19 23:01:47 -05:00
Author
Owner
Copy Link

@cybergod96 commented on GitHub (Jun 3, 2025):

Yep, adding connection through Admin Panel fixes this. Thank you.
I checked documentation again and it clearly tells you to set this up there. When I first launched the app I haven't checked the documentation and tried to set up API using User settings as it was the first thing I noticed. There I noticed that Direct connections is disabled so I enabled it, set up APIs and everything worked fine so I assumed it's the correct way. That said, I think some UI adjustments could be done to prevent this confusion in the future, but for now I think this issue can be closed.

<!-- gh-comment-id:2936135562 --> @cybergod96 commented on GitHub (Jun 3, 2025): Yep, adding connection through Admin Panel fixes this. Thank you. I checked documentation again and it clearly tells you to set this up there. When I first launched the app I haven't checked the documentation and tried to set up API using User settings as it was the first thing I noticed. There I noticed that Direct connections is disabled so I enabled it, set up APIs and everything worked fine so I assumed it's the correct way. That said, I think some UI adjustments could be done to prevent this confusion in the future, but for now I think this issue can be closed.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
confirmed
confirmed issue
core
documentation
enhancement
good first issue
help wanted
non-core
pull-request
Mirrored from GitHub Pull Request
 python
question
testing wanted
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#17298
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?