nisten/mistral-instruct0.2-imatrix4bit.gguf

Base model: https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2

This is just a custom 4bit imatrix quant made to run optiomally on a macbook with 8gb of ram.

For use with llama.cpp https://github.com/ggerganov/llama.cpp