o3-mini returns error 404 from OpenAI API response [REQUIRES UPDATE]

[anphex]

What happened?

Greetings,

I tried to implement o3-mini on my self hosted LibreChat and it returns the error:

"Something went wrong. Here's the specific error message we encountered: Failed to send message. HTTP 404 - { "error": { "message": "This is a chat model and not supported in the v1/completions endpoint. Did you mean to use v1/chat/completions?", "type": "invalid_request_error", "param": "model", "code": null } }"

Weirdly enough, all other models work, even o1.
A little out of scope of this bug, but it would be cool to be able to set the reasoning effort for the latest models through parameters.

Steps to Reproduce

1. Set up LibreChat instance as per documentation.
2. Enable OpenAI in the .env, select models (based on API reference from OpenAI), provide API key and save.
3. Start up the compose
4. All models work, only o3-mini returns an error even though name string for o3-mini is correct. (I also tried the full name of the latest model, same result)

What browsers are you seeing the problem on?

Chrome

Relevant log output

"Something went wrong. Here's the specific error message we encountered: Failed to send message. HTTP 404 - { "error": { "message": "This is a chat model and not supported in the v1/completions endpoint. Did you mean to use v1/chat/completions?", "type": "invalid_request_error", "param": "model", "code": null } }"

Screenshots

Code of Conduct

• I agree to follow this project's Code of Conduct

[danny-avila]

Requires an update.

Updating Instructions (docker):

# Linux command to Remove all existing images
docker images -a | grep "librechat" | awk '{print $3}' | xargs docker rmi

# Windows Powershell
docker images -a | findstr "librechat" | ForEach-Object { docker rmi $_.Split()[2] }

Then follow the directions here: https://www.librechat.ai/docs/local/docker#update-librechat

[larsrinn]

I have the same issue running von version 0.7.7 - is fix not yet in the current release?

[larsrinn]

Ah, I'm seeing this only for o1-pro, which seems to be only available in the responses API