mirror of
https://github.com/open-webui/open-webui.git
synced 2026-05-31 22:08:51 -05:00
[GH-ISSUE #8474] OpenAI: o1-mini doesn't work through openrouter #15138
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/open-webui#15138
Reference in New Issue
Block a user
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?
Originally created by @feronetick on GitHub (Jan 11, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/8474
Installation Method
Installed in LXC container via script from
https://community-scripts.github.io/ProxmoxVE/scripts?id=openwebui
Environment
Confirmation:
Expected Behavior:
o1-mini respond correctly
Actual Behavior:
Endless loading
Description
Bug Summary:
o1-mini model is not working through open router
Reproduction Details
Steps to Reproduce:
Fresh Install latest openwebui in any environment
Add connection to open router: https://openrouter.ai/api/v1
Try to promptOpenAI: o1-mini model
Logs and Screenshots
Browser Console Logs:
[Log] saveSessionSelectedModels (2) (Help.D8aVq3NH.js, line 137)
["openai/o1-mini"] (1)
"["openai/o1-mini"]"
[Log] modelId – "openai/o1-mini" (Help.D8aVq3NH.js, line 138)
[Log] {chat_id: "34a62ab0-56ec-45cb-9bd7-5e43fbffd49e", message_id: "cf289526-9450-45ee-aef0-ec698445d178", data: {type: "chat:completion", data: {done: true, content: "", title: "Новый чат"}}} (Help.D8aVq3NH.js, line 137)
[Log] {status: true, task_id: "45cdad9a-09cf-4a83-b729-becddc6c29e3"} (Help.D8aVq3NH.js, line 142)
[Log] {done: true, content: "", title: "Новый чат"} (Help.D8aVq3NH.js, line 138)
LXC Container Logs:
Jan 11 05:13:54 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "GET /api/v1/users/user/settings HTTP/1.1" 200 OK
Jan 11 05:13:54 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "GET /api/v1/users/user/settings HTTP/1.1" 200 OK
Jan 11 05:14:01 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "POST /api/v1/chats/new HTTP/1.1" 200 OK
Jan 11 05:14:02 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 OK
Jan 11 05:14:02 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "POST /api/v1/chats/34a62ab0-56ec-45cb-9bd7-5e43fbffd49e HTTP/1.1" 200 OK
Jan 11 05:14:02 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 OK
Jan 11 05:14:02 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "POST /api/chat/completions HTTP/1.1" 200 OK
Jan 11 05:14:02 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "POST /api/chat/completed HTTP/1.1" 200 OK
Jan 11 05:14:02 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 OK
Jan 11 05:14:02 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "POST /api/v1/chats/34a62ab0-56ec-45cb-9bd7-5e43fbffd49e HTTP/1.1" 200 OK
Jan 11 05:14:02 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "GET /audio/notification.mp3 HTTP/1.1" 200 OK
Jan 11 05:14:02 openwebui start.sh[9116]: INFO: 192.168.1.1:0 - "GET /api/v1/chats/?page=1 HTTP/1.1" 200 OK
Jan 11 05:14:02 openwebui start.sh[9116]: ERROR [asyncio] Task exception was never retrieved
Jan 11 05:14:02 openwebui start.sh[9116]: future: <Task finished name='Task-237611' coro=<process_chat_response..post_response_handler() done, defined at /opt/open-webui/backend/open_webui/utils/middleware.py:931> exception=IndexError('list index out of range')>
Jan 11 05:14:02 openwebui start.sh[9116]: Traceback (most recent call last):
Jan 11 05:14:02 openwebui start.sh[9116]: File "/opt/open-webui/backend/open_webui/utils/middleware.py", line 1056, in post_response_handler
Jan 11 05:14:02 openwebui start.sh[9116]: await background_tasks_handler()
Jan 11 05:14:02 openwebui start.sh[9116]: File "/opt/open-webui/backend/open_webui/utils/middleware.py", line 774, in background_tasks_handler
Jan 11 05:14:02 openwebui start.sh[9116]: res.get("choices", [])[0]
Jan 11 05:14:02 openwebui start.sh[9116]: ~~~~~~~~~~~~~~~~~~~~~~^^^
Jan 11 05:14:02 openwebui start.sh[9116]: IndexError: list index out of range
Jan 11 05:14:02 openwebui start.sh[9116]: ERROR [asyncio] Unclosed client session
Jan 11 05:14:02 openwebui start.sh[9116]: client_session: <aiohttp.client.ClientSession object at 0x7ba4057b5790>
Jan 11 05:14:02 openwebui start.sh[9116]: ERROR [asyncio] Unclosed connector
Jan 11 05:14:02 openwebui start.sh[9116]: connections: ['deque([(<aiohttp.client_proto.ResponseHandler object at 0x7ba3f81b1c50>, 984258.839415548)])']
Jan 11 05:14:02 openwebui start.sh[9116]: connector: <aiohttp.connector.TCPConnector object at 0x7ba4637d1850>
Screenshots/Screen Recordings (if applicable):
Additional Information
Sometimes I hear the message sound
Note
If the bug report is incomplete or does not follow the provided instructions, it may not be addressed. Please ensure that you have followed the steps outlined in the README.md and troubleshooting.md documents, and provide all necessary information for us to reproduce and address the issue. Thank you!
@tjbck commented on GitHub (Jan 11, 2025):
Should be addressed in dev, testing wanted here!
@rgaricano commented on GitHub (Jan 11, 2025):
Same results, (outside container),
but maybe this error is because I haven't credits for run this model,
when I select a free model (google/gemini-2.0-flash-thinking-exp:free) it responde perfectly.
I did more tests,
after this I removed the model from the OpenRouter connection, logically now I saw the list of all models, I selected o1-mini and continued the chat, I had no error, but I had no text response (all greyed out) (I have no credits),
Ok,
I configure again "openai/o1-mini" model in openrouter conection settings (as I do first time), and I have no errors & no text response (all gray)
I begin a new chat with this model and error come again.
It seem that this error is logged when response (from openrouter in this case) is empty (because no credits for run selected model) & chat is empty.
Maybe it's necessary improve background tasks handle for empty responses & add some emitter regarding this.
@feronetick commented on GitHub (Jan 13, 2025):
So I have enough credits for prompt o1-mini. I can use it in open router web ui. I can share access token for testing purposes if it needed
@rgaricano commented on GitHub (Jan 14, 2025):
new issue reported about: IndexError: list index out of range
@Kovoki commented on GitHub (Jan 14, 2025):
🚀
@tjbck commented on GitHub (Jan 14, 2025):
Closing in favour of https://github.com/open-webui/open-webui/issues/8526