DeepSeek-R1-Distill-Llama-8B-mlx-fp16/README.md

license, library_name, tags, base_model

license	library_name	tags	base_model
mit	transformers	mlx	deepseek-ai/DeepSeek-R1-Distill-Llama-8B

valuat/DeepSeek-R1-Distill-Llama-8B-mlx-fp16

The Model valuat/DeepSeek-R1-Distill-Llama-8B-mlx-fp16 was converted to MLX format from deepseek-ai/DeepSeek-R1-Distill-Llama-8B using mlx-lm version 0.26.4.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("valuat/DeepSeek-R1-Distill-Llama-8B-mlx-fp16")

prompt="hello"

if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)