WaveCut/Qwen_Qwen2.5-14B-Instruct-1M_MLX-4bit

The Model WaveCut/Qwen_Qwen2.5-14B-Instruct-1M_MLX-4bit was converted to MLX format from Qwen/Qwen2.5-14B-Instruct-1M using mlx-lm version 0.21.1.

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("WaveCut/Qwen_Qwen2.5-14B-Instruct-1M_MLX-4bit")

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)