Dolphin3.0-R1-Mistral-24B-MLX-8bits

moot20/Dolphin3.0-R1-Mistral-24B-MLX-8bits

The Model moot20/Dolphin3.0-R1-Mistral-24B-MLX-8bits was converted to MLX format from cognitivecomputations/Dolphin3.0-R1-Mistral-24B using mlx-lm version 0.21.1.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("moot20/Dolphin3.0-R1-Mistral-24B-MLX-8bits")

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)