jay16/deepseek-r1-quantized

The Model jay16/deepseek-r1-quantized was converted to MLX format from mobiuslabsgmbh/DeepSeek-R1-ReDistill-Qwen-1.5B-v1.0 using mlx-lm version 0.21.1.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("jay16/deepseek-r1-quantized")

prompt = "hello"

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

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