How to make localGPT use the local model ?

[50ZAIofficial]

Some HuggingFace models I use do not have a ggml version. I downloaded the model and converted it to model-ggml-q4.bin through llama.cpp, but I cannot call the model through model_id and model_basename. Am I limited to models from the HuggingFace webpage?
Looking forward to your reply, thank you!

[50ZAIofficial]

I modified run_localGPT.py>load_model>
model_path = hf_hub_download(repo_id=model_id, filename=model_basename)
to
model_path = Your-Model-Path-BinFile
like
"C:/Users/X/localGPT/Llama-2-7b-ggml/Llama-2-7b.ggmlv3.q2_K.bin"
it works well

[WarmCatUK]

run_localGPT.py has since changed, and I have the same issue as you. I would like to run a previously downloaded model (mistral-7b-instruct-v0.1.Q8_0.gguf) as I'm currently in a situation where I do not have a fantastic internet connection.
Any ideas?