Using OpenAI omni series models in assistant API

[TheMrCodes]

What happened?

When selecting a o model like o3-mini there is a 400 Server error
because the models don't got an temperature parameter

error: [/assistants/:id] Error updating assistant 400 Unsupported parameter: 'temperature' is not supported with this model.

Version Information

2876993

Steps to Reproduce

Adding o3-mini to ASSISTANTS_MODELS in .env, selecting it in the assistant builder and saving the new assistant

What browsers are you seeing the problem on?

Chrome

Relevant log output

Screenshots

No response

Code of Conduct

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

[danny-avila]

I created this assistant in LC without issues

If you have set temperature in the OpenAI Assistants UI, you need to remove it

[mazierovictor]

@danny-avila

If I use o3-mini as an agent, it works normally, but if I enable tavily to perform searches, it doesn't.

And if I change to GPT-4o, it performs searches normally.

Could it be some LibreChat configuration?