[GH-ISSUE #21331] issue: When using Ollama-compatible LLM provider which does not use ":latest" in its model names no chat is possible #139171

New Issue

GiteaMirror · 2026-05-25T12:37:19-05:00

GiteaMirror commented

2026-05-25 12:37:19 -05:00

Originally created by @daniel-georg369 on GitHub (Feb 12, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/21331

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

0.7.2

Ollama Version (if applicable)

no ollame

Operating System

Windows

Browser (if applicable)

No response

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

An Ollama-compatible LLM provider is added
the models are available to be choosen for a chat
the chat works

Actual Behavior

An Ollama-compatible LLM provider is added
the models are available to be choosen for a chat
the chat does not work, instead the error "400: Model 'azure-o4-mini:latest' was not found".

Steps to Reproduce

Install Windows
Install Docker
Pull and then run the Open Webui container with debug logging
Add Ollama as LLM provider - available models can be chosen for a chat and work
Add a custom LLM provider (which does not use ":xxx" like ":latest" or ":14b" in the model names - available models can be chosen for a chat but don't work

Logs & Screenshots

2026-02-12 15:16:02.437 | DEBUG | open_webui.utils.middleware:process_chat_payload:1641 - tool_ids=None

2026-02-12 15:16:02.437 | DEBUG | open_webui.utils.middleware:process_chat_payload:1642 - direct_tool_servers=[]

2026-02-12 15:16:02.438 | DEBUG | open_webui.utils.chat:generate_chat_completion:171 - generate_chat_completion: {'stream': True, 'model': 'azure-o4-mini', 'messages': [{'role': 'user', '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."}], 'metadata': {'user_id': '50f9c58c-88a6-41df-ad7d-41585fb4a24e', 'chat_id': '01ab8261-b7ee-48cc-91fd-1976321df6f8', 'message_id': '5c397ddc-4c87-4e5c-a936-6fd35fb31b6a', 'parent_message': {'id': 'c0c7598d-41af-4854-9456-887802f0ab2f', 'parentId': None, 'childrenIds': ['5c397ddc-4c87-4e5c-a936-6fd35fb31b6a'], 'role': 'user', '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.", 'timestamp': 1770909362, 'models': ['azure-o4-mini']}, 'parent_message_id': 'c0c7598d-41af-4854-9456-887802f0ab2f', 'session_id': 'VTo6LB43IhJiEhw5AAAD', 'filter_ids': [], 'tool_ids': None, 'tool_servers': [], 'files': None, 'features': {'voice': False, 'image_generation': False, 'code_interpreter': False, 'web_search': False}, 'variables': {'{{USER_NAME}}': 'My User', '{{USER_LOCATION}}': 'Unknown', '{{CURRENT_DATETIME}}': '2026-02-12 16:16:02', '{{CURRENT_DATE}}': '2026-02-12', '{{CURRENT_TIME}}': '16:16:02', '{{CURRENT_WEEKDAY}}': 'Thursday', '{{CURRENT_TIMEZONE}}': 'Europe/Berlin', '{{USER_LANGUAGE}}': 'en-US'}, 'model': {'id': 'azure-o4-mini', 'name': 'azure-o4-mini', 'object': 'model', 'created': 1770909353, 'owned_by': 'ollama', 'ollama': {'name': 'azure-o4-mini', 'model': 'azure-o4-mini', 'connection_type': 'external', 'urls': [2]}, 'connection_type': 'external', 'tags': [], 'actions': [], 'filters': []}, 'direct': False, 'params': {'stream_delta_chunk_size': None, 'reasoning_tags': None, 'function_calling': 'default'}}, 'options': {}}

2026-02-12 15:16:02.440 | DEBUG | open_webui.main:process_chat:1746 - Error processing chat payload: 400: Model 'azure-o4-mini:latest' was not found

Additional Information

The model names delivered by the LLM provider are without ":latest" or ":14b", there is no ":" included in the name.

In main/backend/open_webui/routers/ollama.py there are several code snippets with:
if ":" not in model:
model = f"{model}:latest"

So it seems Open Webui is always adding a ":latest" without it being necessary.
The same LLM provider works without any problems in N8N using their Ollama tool.

Originally created by @daniel-georg369 on GitHub (Feb 12, 2026). Original GitHub issue: https://github.com/open-webui/open-webui/issues/21331 ### 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 0.7.2 ### Ollama Version (if applicable) no ollame ### Operating System Windows ### Browser (if applicable) _No response_ ### 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 - An Ollama-compatible LLM provider is added - the models are available to be choosen for a chat - the chat works ### Actual Behavior - An Ollama-compatible LLM provider is added - the models are available to be choosen for a chat - the chat does not work, instead the error "400: Model 'azure-o4-mini:latest' was not found". ### Steps to Reproduce - Install Windows - Install Docker - Pull and then run the Open Webui container with debug logging - Add Ollama as LLM provider - available models can be chosen for a chat and work - Add a custom LLM provider (which does not use ":xxx" like ":latest" or ":14b" in the model names - available models can be chosen for a chat but don't work - ### Logs & Screenshots 2026-02-12 15:16:02.437 | DEBUG | open_webui.utils.middleware:process_chat_payload:1641 - tool_ids=None 2026-02-12 15:16:02.437 | DEBUG | open_webui.utils.middleware:process_chat_payload:1642 - direct_tool_servers=[] 2026-02-12 15:16:02.438 | DEBUG | open_webui.utils.chat:generate_chat_completion:171 - generate_chat_completion: {'stream': True, 'model': 'azure-o4-mini', 'messages': [{'role': 'user', '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."}], 'metadata': {'user_id': '50f9c58c-88a6-41df-ad7d-41585fb4a24e', 'chat_id': '01ab8261-b7ee-48cc-91fd-1976321df6f8', 'message_id': '5c397ddc-4c87-4e5c-a936-6fd35fb31b6a', 'parent_message': {'id': 'c0c7598d-41af-4854-9456-887802f0ab2f', 'parentId': None, 'childrenIds': ['5c397ddc-4c87-4e5c-a936-6fd35fb31b6a'], 'role': 'user', '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.", 'timestamp': 1770909362, 'models': ['azure-o4-mini']}, 'parent_message_id': 'c0c7598d-41af-4854-9456-887802f0ab2f', 'session_id': 'VTo6LB43IhJiEhw5AAAD', 'filter_ids': [], 'tool_ids': None, 'tool_servers': [], 'files': None, 'features': {'voice': False, 'image_generation': False, 'code_interpreter': False, 'web_search': False}, 'variables': {'{{USER_NAME}}': 'My User', '{{USER_LOCATION}}': 'Unknown', '{{CURRENT_DATETIME}}': '2026-02-12 16:16:02', '{{CURRENT_DATE}}': '2026-02-12', '{{CURRENT_TIME}}': '16:16:02', '{{CURRENT_WEEKDAY}}': 'Thursday', '{{CURRENT_TIMEZONE}}': 'Europe/Berlin', '{{USER_LANGUAGE}}': 'en-US'}, 'model': {'id': 'azure-o4-mini', 'name': 'azure-o4-mini', 'object': 'model', 'created': 1770909353, 'owned_by': 'ollama', 'ollama': {'name': 'azure-o4-mini', 'model': 'azure-o4-mini', 'connection_type': 'external', 'urls': [2]}, 'connection_type': 'external', 'tags': [], 'actions': [], 'filters': []}, 'direct': False, 'params': {'stream_delta_chunk_size': None, 'reasoning_tags': None, 'function_calling': 'default'}}, 'options': {}} 2026-02-12 15:16:02.440 | DEBUG | open_webui.main:process_chat:1746 - Error processing chat payload: 400: Model 'azure-o4-mini:latest' was not found <img width="1409" height="327" alt="Image" src="https://github.com/user-attachments/assets/27a4004d-e5db-4068-a680-2e7faf03b4b0" /> ### Additional Information The model names delivered by the LLM provider are without ":latest" or ":14b", there is no ":" included in the name. In main/backend/open_webui/routers/ollama.py there are several code snippets with: if ":" not in model: model = f"{model}:latest" So it seems Open Webui is always adding a ":latest" without it being necessary. The same LLM provider works without any problems in N8N using their Ollama tool.

GiteaMirror added the bug label 2026-05-25 12:37:19 -05:00

GiteaMirror closed this issue

2026-05-25 12:37:20 -05:00

GiteaMirror commented

2026-05-25 12:37:22 -05:00

@Classic298 commented on GitHub (Feb 12, 2026):

fixed by 05ae44b98d

@Classic298 commented on GitHub (Feb 12, 2026): fixed by https://github.com/open-webui/open-webui/commit/05ae44b98dc279ee12cc8eab17278ccbfec60301

GiteaMirror commented

2026-05-25 12:37:22 -05:00

@manuel-koch commented on GitHub (Feb 13, 2026):

I updated to v0.8.0 but still ollama cloud models that don't use ":latest" tag can't be used in Open WebUI.
I removed such a model from my ollama provider configuration, saved the config and re-added the model ( kimi-k2-thinking ).
Afterwards trying to chat with that model always returns "400: Model 'cloud.kimi-k2-thinking:latest' was not found".

The model exists on ollama cloud:

curl https://ollama.com/api/chat \
  -H "Authorization: Bearer $OLLAMA_API_KEY" \
  -d '{
    "model": "kimi-k2-thinking",
    "messages": [{
      "role": "user",
      "content": "Why is the sky blue?"
    }],
    "stream": false
  }'

# returns a chat response

@Classic298 Anything else I need to do to get it working ?

I checked my Open WebUI configuration and I'm using MODELS_CACHE_TTL="1200" too.
But AFAIK editing a provider or a model invalidates the cached models.

I have edited the model config for kimi-k2-thinking. Even if I remove the model from the ollama provider and save/reload and re-add it afterwards and save/reload my edited model config for kimi-k2-thinking is re-used - could that be a problem ?

Any cache that needs to be invalidated ? How ?

@manuel-koch commented on GitHub (Feb 13, 2026): I updated to v0.8.0 but still ollama cloud models that don't use ":latest" tag can't be used in Open WebUI. I removed such a model from my ollama provider configuration, saved the config and re-added the model ( kimi-k2-thinking ). Afterwards trying to chat with that model always returns "400: Model 'cloud.kimi-k2-thinking:latest' was not found". The model exists on ollama cloud: ```shell curl https://ollama.com/api/chat \ -H "Authorization: Bearer $OLLAMA_API_KEY" \ -d '{ "model": "kimi-k2-thinking", "messages": [{ "role": "user", "content": "Why is the sky blue?" }], "stream": false }' # returns a chat response ``` @Classic298 Anything else I need to do to get it working ? <del>I checked my Open WebUI configuration and I'm using `MODELS_CACHE_TTL="1200"` too. But AFAIK editing a provider or a model invalidates the cached models.</del> I have edited the model config for kimi-k2-thinking. Even if I remove the model from the ollama provider and save/reload and re-add it afterwards and save/reload my edited model config for kimi-k2-thinking is re-used - could that be a problem ? Any cache that needs to be invalidated ? How ?

GiteaMirror commented

2026-05-25 12:37:23 -05:00

@daniel-georg369 commented on GitHub (Feb 13, 2026):

I also just updated to 0.8.0, for us the problem is fixed. We now see our models as "modelname":"providername"
Thanks for the extremely fast fix! :)

@daniel-georg369 commented on GitHub (Feb 13, 2026): I also just updated to 0.8.0, for us the problem is fixed. We now see our models as "modelname":"providername" Thanks for the extremely fast fix! :)

GiteaMirror commented

2026-05-25 12:37:24 -05:00

@manuel-koch commented on GitHub (Feb 13, 2026):

I'm running another Open WebUI instance where no Ollama Cloud provider was used before.
Adding a ollama cloud provider ( incl. the model kimi-k2-thinking ), save/reload, check that model appears in the list of models, try to chat: "400: Model 'cloud.kimi-k2-thinking:latest' was not found"

@manuel-koch commented on GitHub (Feb 13, 2026): I'm running another Open WebUI instance where no Ollama Cloud provider was used before. Adding a ollama cloud provider ( incl. the model kimi-k2-thinking ), save/reload, check that model appears in the list of models, try to chat: "400: Model 'cloud.kimi-k2-thinking:latest' was not found"

GiteaMirror commented

2026-05-25 12:37:25 -05:00

@manuel-koch commented on GitHub (Feb 13, 2026):

See a related discussion.

@manuel-koch commented on GitHub (Feb 13, 2026): See a related [discussion](https://github.com/open-webui/open-webui/discussions/19142).

Sign in to join this conversation.

Branches Tags

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/open-webui#139171