mirror of
https://github.com/open-webui/open-webui.git
synced 2026-07-11 10:34:13 -05:00
[GH-ISSUE #15662] issue: /models called spammly
#17630
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/open-webui#17630
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 @steveepreston on GitHub (Jul 11, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/15662
Check Existing Issues
Installation Method
Docker
Open WebUI Version
0.6.15
Ollama Version (if applicable)
No response
Operating System
Windows 11
Browser (if applicable)
Chrome
Confirmation
README.md.Expected Behavior
Open-webui spamming
/modelscall in chat screen, on selecting model, or when a model responsefor example about 4-5 calls in just 1 second
Actual Behavior
call the
/modelsnormallySteps to Reproduce
Open-webui start calling
/modelsmultiple timesOpen-webui start calling
/modelsmultiple timesLogs & Screenshots
Docker Logs:
Browser Network logs:
Additional Information
Suggestedsection in chat start screen is flashing due to high speed repetitive/modelscalls@steveepreston commented on GitHub (Jul 11, 2025):
Why this bug is critical?
Because when it start spamming
/modelscall, on a request between all requests will get an error (like 429 rate limiting) and it causes to models list on top-left be cleared and page break-down.@tjbck commented on GitHub (Jul 11, 2025):
This is not a bug and is an intended behaviour to ensure the model list is synced with the backend.
@steveepreston commented on GitHub (Jul 12, 2025):
@tjbck no, this misbehaving
as i logged, this break the page, and models list get 429 and cant load models anymore
@steveepreston commented on GitHub (Jul 12, 2025):
@tjbck is it normally to call
/models12 times at a moment?look at the requestes timeline, all requests initiated at once
i cleared the network request history, then clicked on
New Chatand took this screenshotman this clearly is a bug
@rgaricano commented on GitHub (Jul 12, 2025):
This could be by "error" in db, and the fact that this call haven't exceptions check (try)
I think that it's a particular issue that can arise due to a bad/misconfiguration or error in db.
And a exception control should give more concrete info about the problem.
(I can't reproduce, I have just one models fetch call)
5eca495d3e/backend/open_webui/utils/chat.py (L291)5eca495d3e/backend/open_webui/utils/models.py (L80)5eca495d3e/backend/open_webui/models/models.py (L173)5eca495d3e/backend/open_webui/routers/openai.py (L396)Also a openwebui log could help to find the problem.
@rgaricano commented on GitHub (Jul 12, 2025):
Steve, could you provide log when this happend?
@steveepreston commented on GitHub (Jul 12, 2025):
@rgaricano Thanks for attention 🙏
it's really broking the page, because it makes models list
emptyand should refresh to continue@steveepreston commented on GitHub (Jul 12, 2025):
@rgaricano yes man, please tell me where to watch for logs?
i screenshoted the
Docker logsandNetwork tab requestsat first message@rgaricano commented on GitHub (Jul 12, 2025):
Yes, that it's the issue, it can't get correctly the model list and it retry,
it could be by db "error" or an error in connection (please, recheck connection url & auth).
For find it it's necessary look at log (in debug mode)
@rgaricano commented on GitHub (Jul 12, 2025):
run it with env var: GLOBAL_LOG_LEVEL="DEBUG"
@tjbck commented on GitHub (Jul 12, 2025):
@steveepreston as i've said, when you hover over the model list it'll make a network request to fetch the model list to make sure you're up to date with the latest model list. This on its own NOT an issue. However, the UI not working does seem like a some sort of an issue, but with the information provided we cannot reproduce the issue.