AI Agent model input suggestion

The idea is:

HTTP request to connect to a custom LLM model as input for AI agent

My use case:

I’ve been using some 3rd party LLM which is not currently on the LLM mode list and its not something i can run locally (i.e. cant use the AI starter kit) that I’m calling via HTTP request, would be nice if the AI agent could take HTTP request to these LLM models as input

I think it would be beneficial to add this because:

It opens many more doors for me to test with different LLMs

Any resources to support this?

e.g. Perplexity AI, and many more

Are you willing to work on this?

Yes