mistralai/mistral-inference ? reverse-engineered prompt
Reverse engineered prompt
Build me a small Python tool for running Mistral models locally.
I want it to let me download a model folder, then run a quick demo command to see if everything works, and a chat command so I can talk to the model from the terminal. It should support the main Mistral open weight models mentioned in the docs, including the smaller ones that can run on one GPU and the bigger ones that need multiple GPUs. Please make it easy to use with a simple install, clear setup steps, and examples for both local model files and Hugging Face downloads.
If you need to check current model or download details, look up the latest docs online. Also include a few basic tutorials or notebooks that show how to get started and how to call the model from Python.
Are you gonna build this?
make sure you review the code using coderabbit