[Bug]: supports_httpx_timeout is too restrictive

[npt]

What happened?

Currently, if the provider is anything but openai, azure, or bedrock, an httpx.Timeout passed in will be turned into a float. This can be less restrictive — when I monkeypatch supports_httpx_timeout to allow through custom_openai as well, it seems to work fine. supports_httpx_timeout should also allow custom_openai, and presumably other providers that use the OpenAI client.

Relevant log output

Are you a ML Ops Team?

No

What LiteLLM version are you on ?

v1.60.6

Twitter / LinkedIn details

No response