[GH-ISSUE #22456] issue: Custom Model Fallback Unreachable #90456

Closed
opened 2026-05-15 15:42:51 -05:00 by GiteaMirror · 2 comments
Owner

Originally created by @castielwaverly on GitHub (Mar 8, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/22456

Check Existing Issues

  • I have searched for any existing and/or related issues.
  • I have searched for any existing and/or related discussions.
  • I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!).
  • I am using the latest version of Open WebUI.

Installation Method

Docker

Open WebUI Version

v0.8.9

Ollama Version (if applicable)

v0.17.4

Operating System

Alpine 3.23

Browser (if applicable)

Firefox

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

When I have enabled fallbacks and specified a default fallback model, the fallback gets used when the current base model of my custom model is unavailable.

Actual Behavior

When I have enabled fallbacks and specified a default fallback model, I get a 404: Model not found error when the current base model of my custom model is unavailable.

Steps to Reproduce

  1. Set up two ollama instances, e.g. ollama-a and ollama-b
  2. Create a custom model that uses a base model from ollama-a, e.g. ollama-a/gpt-oss:20b
  3. Set the default model to a model from ollama-b, e.g. ollama-b/ministral-3:3b
  4. Chat with the custom model and observe it is using ollama-a/gpt-oss:20b correctly.
  5. Take down ollama-a instance.
  6. Chat with custom model and observe the error

Logs & Screenshots

open_webui.main:process_chat:1889 - Error processing chat payload: 404: Model not found

Additional Information

This is happening because model_info_params.get("base_model_id") seems to be missing in all situations (instance down or not), so this entire block gets skipped:

https://github.com/open-webui/open-webui/blob/6c159a97b7efdfdbfa262ebd26823a2d695b561d/backend/open_webui/main.py#L1734-L1741

This model info comes from the global config as the base, with specifics overwritten by the model params
https://github.com/open-webui/open-webui/blob/6c159a97b7efdfdbfa262ebd26823a2d695b561d/backend/open_webui/main.py#L1721-L1732

While base_model_id is part of ModelModel, it's not part of the params that are being used there it seems
https://github.com/open-webui/open-webui/blob/6c159a97b7efdfdbfa262ebd26823a2d695b561d/backend/open_webui/models/models.py#L86-L89

As a consequence, model and form_data['model'] stay as their current values and not the fallback model, which fails downstream in the pipeline. I also experimented with forcing that block to run by manipulating the code directly on my deployment, and it does proceed to processing the chat, but fails somewhere else downstream more silently -- perhaps for reasons related to the original problem described here happening elsewhere.

Originally created by @castielwaverly on GitHub (Mar 8, 2026). Original GitHub issue: https://github.com/open-webui/open-webui/issues/22456 ### Check Existing Issues - [x] I have searched for any existing and/or related issues. - [x] I have searched for any existing and/or related discussions. - [x] I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!). - [x] I am using the latest version of Open WebUI. ### Installation Method Docker ### Open WebUI Version v0.8.9 ### Ollama Version (if applicable) v0.17.4 ### Operating System Alpine 3.23 ### Browser (if applicable) Firefox ### 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 When I have enabled fallbacks and specified a default fallback model, the fallback gets used when the current base model of my custom model is unavailable. ### Actual Behavior When I have enabled fallbacks and specified a default fallback model, I get a `404: Model not found` error when the current base model of my custom model is unavailable. ### Steps to Reproduce 1. Set up two ollama instances, e.g. `ollama-a` and `ollama-b` 2. Create a custom model that uses a base model from `ollama-a`, e.g. `ollama-a/gpt-oss:20b` 3. Set the default model to a model from `ollama-b`, e.g. `ollama-b/ministral-3:3b` 4. Chat with the custom model and observe it is using `ollama-a/gpt-oss:20b` correctly. 5. Take down `ollama-a` instance. 6. Chat with custom model and observe the error ### Logs & Screenshots ``` open_webui.main:process_chat:1889 - Error processing chat payload: 404: Model not found ``` ### Additional Information This is happening because `model_info_params.get("base_model_id")` seems to be missing in all situations (instance down or not), so this entire block gets skipped: https://github.com/open-webui/open-webui/blob/6c159a97b7efdfdbfa262ebd26823a2d695b561d/backend/open_webui/main.py#L1734-L1741 This model info comes from the global config as the base, with specifics overwritten by the model params https://github.com/open-webui/open-webui/blob/6c159a97b7efdfdbfa262ebd26823a2d695b561d/backend/open_webui/main.py#L1721-L1732 While `base_model_id` is part of `ModelModel`, it's not part of the params that are being used there it seems https://github.com/open-webui/open-webui/blob/6c159a97b7efdfdbfa262ebd26823a2d695b561d/backend/open_webui/models/models.py#L86-L89 As a consequence, `model` and `form_data['model']` stay as their current values and not the fallback model, which fails downstream in the pipeline. I also experimented with forcing that block to run by manipulating the code directly on my deployment, and it does proceed to processing the chat, but fails somewhere else downstream more silently -- perhaps for reasons related to the original problem described here happening elsewhere.
GiteaMirror added the bug label 2026-05-15 15:42:51 -05:00
Author
Owner

@tjbck commented on GitHub (Mar 8, 2026):

Addressed in dev.

<!-- gh-comment-id:4020154598 --> @tjbck commented on GitHub (Mar 8, 2026): Addressed in dev.
Author
Owner
<!-- gh-comment-id:4020156754 --> @Classic298 commented on GitHub (Mar 8, 2026): https://github.com/open-webui/open-webui/commit/c97767424f6ac7daf61daa84e15cdb10f1965d36
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#90456