dfurman/Mistral-7B-Instruct-v0.3-mlx-4bit

The Model dfurman/Mistral-7B-Instruct-v0.3-mlx-4bit was converted to MLX format from mistralai/Mistral-7B-Instruct-v0.3 using mlx-lm version 0.14.0.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("dfurman/Mistral-7B-Instruct-v0.3-mlx-4bit")
response = generate(model, tokenizer, prompt="hello", verbose=True)