Assistant model display error

[pxz2016]

Summary

The model of the assistant found by the API has been modified to the first model of the endpoint, resulting in a discrepancy between the model displayed on the UI and the actual model of the assistant.

Change Type

Please delete any irrelevant options.

• Bug fix (non-breaking change which fixes an issue)

Checklist

Please delete any irrelevant options.

• My code adheres to this project's style guidelines
• I have performed a self-review of my own code
• I have commented in any complex areas of my code
• I have made pertinent documentation changes
• My changes do not introduce new warnings
• I have written tests demonstrating that my changes are effective or that my feature works
• Local unit tests pass with my changes
• Any changes dependent on mine have been merged and published in downstream modules.
• New documents have been locally validated with mkdocs

[danny-avila]

thank you! Will investigate a before/after of this

[danny-avila]

With your changes:

This is a tricky setup because the intention here is allow multiple groups. The model names with a librechat.yaml setup do not have to be the same as the deployment name, but Azure lists the models as deployment names. With just the deployment names, we can't map to the appropriate group because a group can have the same deployment name as another, but they cannot share model names, which is why I'm using the model name.

Thank you for the PR though which highlights the issue.

[danny-avila]

and to clarify I will think about a few different things we can do here. one additional step to mitigate this is to check if the deployment name is specified at the model-level; however, if it's not, the first model of the group will be used