Custom

@woojim just wrap the api into a custom llm object.

If the api supports message roles, you can also implement the chat/stream_chat functions

https://gpt-index.readthedocs.io/en/stable/core_modules/model_modules/llms/usage_custom.html#example-using-a-custom-llm-model-advanced

Oh thanks for the link! That's what I was thinking and was wondering about.