Qwen3.5-27B-GLM5.1-Distill-v1-qx86-hi-mlx

Brainwaves

          arc   arc/e boolq hswag obkqa piqa  wino
qx86-hi   0.451,0.497,0.740,0.716,0.390,0.778,0.749

Instruct
qx86-hi   0.619,0.775,0.900,0.735,0.440,0.801,0.713

Quant     Perplexity      Peak Memory   Tokens/sec
qx86-hi   3.836 ± 0.024   32.36 GB      135

Baseline model

Qwen3.5-27B
          arc   arc/e boolq hswag obkqa piqa  wino
qx86-hi   0.443,0.498,0.857,0.701,0.372,0.770,0.752

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("Qwen3.5-27B-GLM5.1-Distill-v1-qx86-hi-mlx")

prompt = "hello"

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

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