[Feature]: add the ability for the local vLLM provider to select a provider
If me use this project https://github.com/xtekky/gpt4free We can use the compatible api at http://localhost:8080/v1, but there is an option to choose the provider there. Which we can't do in the current implementation. also there is a problem the first message in the chat goes away, but the rest do not: Full Error: { "status": 422, "headers": { "content-length": "562", "content-type": "application/json", "date": "Sun, 18 May 2025 05:18:52 GMT", "server": "uvicorn" } } I don't know what other pitfalls there might be.
Anyone want to implement this? Should be simple to add a new vLLM field, Claude 4 can probably one shot this! Will revisit this eventually if no one does a PR before I do.
should I do it ?
Sure @epiitom! Happy to help or answer any questions.
Is there any update on this @andrewpareles @epiitom.