Cannot run open interpreter with a local llamfile

[RoseDeSable]

Describe the bug

Hi,
I run the interpreter under a debian linux (KaliLinux) in a virtual environment. I start it with the command "interpreter --model rocket-3b.Q4_K_M.llamafile". But the interpreter alyways want to use it with the protocol of openai. I don't understand this, because I installed the interface "llama-cpp-python". If I start the model seperately, then I can chat with it by my firefox browser. Must I start the model at first and then start the interpreter with the base_api option ? I mean the command "interpreter --base_api http://localhost:port(of the model)".

Best Regards
Rose

Reproduce

Yes

Expected behavior

I believe, that the interpreter start the model file in the background and chats by lama-cpp-python with it. So like firefox does.

Screenshots

No response

Open Interpreter version

Version: 0.4.3

Python version

Python 3.11.9

Operating System name and version

Debian

Additional context

[RoseDeSable]

Ok,
after regarding this short video https://www.youtube.com/watch?v=MYmW786xCRE, I started first my model. After it is initialized, I can chat with it in my browser. Now I start the interpreter:

interpreter --api_base http://127.0.0.1:8080/v1 --model rocket
-3b.Q4_K_M.llamafile --api_key dummykey

and perfect, it goes

▌ Model set to openai/rocket-3b.Q4_K_M.llamafile

Open Interpreter will require approval before running code.

Use interpreter -y to bypass this.

Press CTRL-C to exit.

> Hello

We were unable to determine the context window of this model. Defaulting to 8000.

If your model can handle more, run interpreter --context_window {token limit} --max_tokens {max tokens per response}.

Continuing...


  Hello! I am an AI language model and I can assist you with a variety of tasks, answer questions, provide
  information, and more. What would you like me to do for you today? Just let me know and I'll do my best to help.

>

Bye

now my problem is solved