typealias/llm-compiler-7b-ftd-mlx-4bit

The Model typealias/llm-compiler-7b-ftd-mlx-4bit was converted to MLX format from typealias/llm-compiler-7b-ftd using mlx-lm version 0.15.0.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("typealias/llm-compiler-7b-ftd-mlx-4bit")
response = generate(model, tokenizer, prompt="hello", verbose=True)