Firstly, thank you for all your hard work on the Open-webui package. I appreciate that it's provided for free and has been very useful.
I'm experiencing an issue where the package seems to fetch the list of models from the API every time a message is sent. This often leads to connection errors, which interrupts the workflow. I understand there might be reasons for fetching the model list frequently, but I wanted to bring this to your attention in case there's room for optimization.
Installation Method:
Installed via PIP
Open-webui version: 0.4.8
Environment:
Operating System: Windows 11
Expected Behavior:
The package would fetch the model list once and reuse it for subsequent messages, minimizing unnecessary API calls and reducing the likelihood of connection errors.
Actual Behavior:
With every message sent, the package fetches the model list from the API.
This results in frequent connection errors, even after increasing the timeout settings for OpenAI models.
Logs and Screenshots:
Error messages received after sending messages:
Additional Information:
Increasing the timeout for OpenAI models did not resolve the connection errors.
I haven't found similar issues reported by others, so I'm not sure if this is an isolated case or not.
I'm not certain of the specific steps to reproduce the issue consistently, but it occurs frequently during normal use.
Suggestions:
If possible, implementing a caching mechanism for the model list might help improve performance and reduce connection errors.
Alternatively, providing an option to manually refresh the model list or to set automatic refresh intervals could be beneficial.
Handling connection errors gracefully to prevent disruptions in usage would enhance the user experience.
Again, thank you for your time and effort in developing and maintaining this package. Any guidance or assistance in resolving this issue would be greatly appreciated.
Originally created by @Adhithya03 on GitHub (Dec 24, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/8039
Hello,
Firstly, thank you for all your hard work on the Open-webui package. I appreciate that it's provided for free and has been very useful.
I'm experiencing an issue where the package seems to fetch the list of models from the API every time a message is sent. This often leads to connection errors, which interrupts the workflow. I understand there might be reasons for fetching the model list frequently, but I wanted to bring this to your attention in case there's room for optimization.
**Installation Method:**
- Installed via PIP
- Open-webui version: 0.4.8
**Environment:**
- Operating System: Windows 11
**Expected Behavior:**
- The package would fetch the model list once and reuse it for subsequent messages, minimizing unnecessary API calls and reducing the likelihood of connection errors.
**Actual Behavior:**
- With every message sent, the package fetches the model list from the API.
- This results in frequent connection errors, even after increasing the timeout settings for OpenAI models.
**Logs and Screenshots:**
- Error messages received after sending messages:
**Additional Information:**
- Increasing the timeout for OpenAI models did not resolve the connection errors.
- I haven't found similar issues reported by others, so I'm not sure if this is an isolated case or not.
- I'm not certain of the specific steps to reproduce the issue consistently, but it occurs frequently during normal use.
**Suggestions:**
- If possible, implementing a caching mechanism for the model list might help improve performance and reduce connection errors.
- Alternatively, providing an option to manually refresh the model list or to set automatic refresh intervals could be beneficial.
- Handling connection errors gracefully to prevent disruptions in usage would enhance the user experience.
---
Again, thank you for your time and effort in developing and maintaining this package. Any guidance or assistance in resolving this issue would be greatly appreciated.
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 @Adhithya03 on GitHub (Dec 24, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/8039
Hello,
Firstly, thank you for all your hard work on the Open-webui package. I appreciate that it's provided for free and has been very useful.
I'm experiencing an issue where the package seems to fetch the list of models from the API every time a message is sent. This often leads to connection errors, which interrupts the workflow. I understand there might be reasons for fetching the model list frequently, but I wanted to bring this to your attention in case there's room for optimization.
Installation Method:
Environment:
Expected Behavior:
Actual Behavior:
Logs and Screenshots:
Error messages received after sending messages:
Additional Information:
Suggestions:
Again, thank you for your time and effort in developing and maintaining this package. Any guidance or assistance in resolving this issue would be greatly appreciated.