I have searched the existing issues and discussions.
I am using the latest version of Open WebUI.
Installation Method
Docker
Open WebUI Version
0.6.15
Ollama Version (if applicable)
No response
Operating System
Windows 11
Browser (if applicable)
Chrome
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
Open-webui spamming /models call in chat screen, on selecting model, or when a model response
for example about 4-5 calls in just 1 second
Actual Behavior
call the /models normally
Steps to Reproduce
Run Windows 11
Install Docker
Pull latest image
Run Open-webui
Add Model Endpoint
Open chat
Open models list to select models Open-webui start calling /models multiple times
Send a message to models
Models trying to show response Open-webui start calling /models multiple times
Logs & Screenshots
Docker Logs:
Browser Network logs:
Additional Information
Note: Tested on a new installed browser and issue exists
Note: the Suggested section in chat start screen is flashing due to high speed repetitive /models calls
Note: Recreating container didn't fixed
Originally created by @steveepreston on GitHub (Jul 11, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/15662
### Check Existing Issues
- [x] I have searched the existing issues and discussions.
- [x] I am using the latest version of Open WebUI.
### Installation Method
Docker
### Open WebUI Version
0.6.15
### Ollama Version (if applicable)
_No response_
### Operating System
Windows 11
### Browser (if applicable)
Chrome
### 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
Open-webui spamming `/models` call in chat screen, on selecting model, or when a model response
for example about 4-5 calls in just 1 second
### Actual Behavior
call the `/models` normally
### Steps to Reproduce
1. Run Windows 11
2. Install Docker
3. Pull latest image
4. Run Open-webui
5. Add Model Endpoint
6. Open chat
7. Open models list to select models
*Open-webui start calling `/models` multiple times*
8. Send a message to models
9. Models trying to show response
*Open-webui start calling `/models` multiple times*
### Logs & Screenshots
**Docker Logs:**
<img width="1092" height="1069" alt="Image" src="https://github.com/user-attachments/assets/959b7450-a06d-4fd5-8ba9-081e1c0cffc9" />
**Browser Network logs:**
<img width="2318" height="998" alt="Image" src="https://github.com/user-attachments/assets/1ee8b528-a1d2-4e47-b73b-72975840bc40" />
### Additional Information
- **Note:** Tested on a new installed browser and issue exists
- **Note:** the `Suggested` section in chat start screen is flashing due to high speed repetitive `/models` calls
- **Note:** Recreating container didn't fixed
GiteaMirror
added the bug label 2026-04-19 23:26:11 -05:00
@steveepreston commented on GitHub (Jul 11, 2025):
Why this bug is critical?
Because when it start spamming /models call, 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.
<!-- gh-comment-id:3063928179 -->
@steveepreston commented on GitHub (Jul 11, 2025):
Why this bug is critical?
Because when it start spamming `/models` call, 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.
This is not a bug and is an intended behaviour to ensure the model list is synced with the backend.
<!-- gh-comment-id:3063970098 -->
@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
<!-- gh-comment-id:3064684573 -->
@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 /models 12 times at a moment?
look at the requestes timeline, all requests initiated at once
i cleared the network request history, then clicked on New Chat and took this screenshot
man this clearly is a bug
<!-- gh-comment-id:3064691288 -->
@steveepreston commented on GitHub (Jul 12, 2025):
@tjbck is it normally to call `/models` 12 times at a moment?
look at the requestes timeline, all requests initiated at once
<img width="1646" height="505" alt="Image" src="https://github.com/user-attachments/assets/b2156f70-7556-4512-a227-616b150553d9" />
i cleared the network request history, then clicked on `New Chat` and took this screenshot
man this clearly is a bug
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)
Also a openwebui log could help to find the problem.
<!-- gh-comment-id:3064937168 -->
@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)
https://github.com/open-webui/open-webui/blob/5eca495d3e3b3066e7831141ed2adffbd6d179b4/backend/open_webui/utils/chat.py#L291
https://github.com/open-webui/open-webui/blob/5eca495d3e3b3066e7831141ed2adffbd6d179b4/backend/open_webui/utils/models.py#L80
https://github.com/open-webui/open-webui/blob/5eca495d3e3b3066e7831141ed2adffbd6d179b4/backend/open_webui/models/models.py#L173
https://github.com/open-webui/open-webui/blob/5eca495d3e3b3066e7831141ed2adffbd6d179b4/backend/open_webui/routers/openai.py#L396
Also a openwebui log could help to find the problem.
@steveepreston commented on GitHub (Jul 12, 2025):
@rgaricano Thanks for attention 🙏
it's really broking the page, because it makes models list empty and should refresh to continue
<!-- gh-comment-id:3064948087 -->
@steveepreston commented on GitHub (Jul 12, 2025):
@rgaricano Thanks for attention 🙏
it's really broking the page, because it makes models list `empty` and should refresh to continue
<img width="376" height="314" alt="Image" src="https://github.com/user-attachments/assets/39705e11-3275-43cf-8707-02bd63d90dbd" />
@steveepreston commented on GitHub (Jul 12, 2025):
@rgaricano yes man, please tell me where to watch for logs?
i screenshoted the Docker logs and Network tab requests at first message
<!-- gh-comment-id:3064949927 -->
@steveepreston commented on GitHub (Jul 12, 2025):
@rgaricano yes man, please tell me where to watch for logs?
i screenshoted the `Docker logs` and `Network tab requests` at first message
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)
<!-- gh-comment-id:3064950011 -->
@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 yes man, please tell me where to watch for logs? i screenshoted the Docker logs and Network tab requests at first message
run it with env var: GLOBAL_LOG_LEVEL="DEBUG"
<!-- gh-comment-id:3064951180 -->
@rgaricano commented on GitHub (Jul 12, 2025):
> [@rgaricano](https://github.com/rgaricano) yes man, please tell me where to watch for logs? i screenshoted the `Docker logs` and `Network tab requests` at first message
run it with env var: GLOBAL_LOG_LEVEL="DEBUG"
@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.
<!-- gh-comment-id:3064953419 -->
@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.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
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)
https://github.com/open-webui/open-webui/blob/5eca495d3e3b3066e7831141ed2adffbd6d179b4/backend/open_webui/utils/chat.py#L291
https://github.com/open-webui/open-webui/blob/5eca495d3e3b3066e7831141ed2adffbd6d179b4/backend/open_webui/utils/models.py#L80
https://github.com/open-webui/open-webui/blob/5eca495d3e3b3066e7831141ed2adffbd6d179b4/backend/open_webui/models/models.py#L173
https://github.com/open-webui/open-webui/blob/5eca495d3e3b3066e7831141ed2adffbd6d179b4/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.