Support for Custom Backends in Composer

[anthonys]

I have followed the thread about local hosting models and in there is mentioned that custom model backend don’t support Composer or autocomplete.

Using Local LLMs with Cursor: Is it Possible?

I don’t see any good reason why this should not be doable, or made doable. Any RAG or prompt optimization can be made client-side and the cache either made local or cut entirely. We should be able to use this on international flights, which is not possible right now.

Is there any plans to support Composer and Autocomplete with custom backend’s instead of just with claude and gpt models?