Modelpublic

sdkv2/falcon-h1-1.5b-mlx-4bit

sourceHugging Faceupdated 25d agoView on Hugging Face

0likes154downloads

Model Card

Falcon-H1 1.5B MLX 4-bit

4-bit quantized MLX port of tiiuae/Falcon-H1-1.5B-Base for Apple Silicon.

For BF16: `sdkv2/falcon-h1-1.5b-mlx-bf16`

Usage

python

from mlx_lm import load, generate

model, tokenizer = load("sdkv2/falcon-h1-1.5b-mlx-4bit")
response = generate(model, tokenizer, prompt="Hello", verbose=True)