Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled-4bit-MLX

pip install mlx-lm

python -m mlx_lm.chat \
  --model BeastCode/Qwen3.5-27B-Claude-4.6-Opus-Distilled-MLX-4bit

from mlx_lm import load, generate

model, tokenizer = load("BeastCode/Qwen3.5-27B-Claude-4.6-Opus-Distilled-MLX-4bit")

messages = [
    {
        "role": "user",
        "content": (
            "A farmer needs to cross a river with a wolf, a goat, and a cabbage. "
            "The boat can only hold the farmer and one other item. "
            "If left alone, the wolf eats the goat, and the goat eats the cabbage. "
            "How can he get everything across safely?"
        ),
    }
]

# enable_thinking=True inserts the <think> prefix automatically
prompt = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=True,
)

response = generate(model, tokenizer, prompt=prompt, max_tokens=2048, verbose=True)
print(response)

prompt = (
    "<|im_start|>system\n"
    "You are a highly analytical assistant.\n"
    "<|im_end|>\n"
    "<|im_start|>user\n"
    "A farmer needs to cross a river with a wolf, a goat, and a cabbage. "
    "The boat can only hold the farmer and one other item. "
    "If left alone, the wolf eats the goat, and the goat eats the cabbage. "
    "How can he get everything across safely?\n"
    "<|im_end|>\n"
    "<|im_start|>assistant\n"
    "<think>\n"
)

response = generate(model, tokenizer, prompt=prompt, max_tokens=2048, verbose=True)
print(response)

python -m mlx_lm.convert \
  --hf-path Jackrong/Qwen3.5-27B-Claude-4.6-Opus-Reasoning-Distilled \
  --mlx-path ./Qwen3.5-27B-Claude-4.6-MLX \
  --quantize \
  --q-bits 4

import re

def strip_thinking(text: str) -> str:
    """Remove the internal <think> process, returning only the final answer."""
    return re.sub(r'<think>.*?</think>\s*', '', text, flags=re.DOTALL).strip()