Feature Request: support for llama.cpp webserver multiple models

[userman2213]

Would be great to have a support of the llama.cpp webserver api, is a openAI dropin api. And then also to support multiple models on the llama.cpp.
i use llama.cpp in my local network and configured it with more than 1 model. so it would be cool if its possible to easy switch these models from the gui like its right now between openai and groq