Qwen3.6-27B-Claude-4.6-OS-mxfp8-mlx

Brainwaves (Instruct)

         arc   arc/e boolq hswag obkqa piqa  wino
bf16     0.683,0.858,0.910,0.797,0.494,0.820,0.755
mxfp8    0.695,0.869,0.910,0.791,0.504,0.824,0.760
qx64-hi  0.688,0.859,0.903

Quant    Perplexity      Peak Memory   Tokens/sec
mxfp8    4.006 ± 0.026   34.74 GB      187  
qx64-hi  4.098 ± 0.027   25.64 GB      208

Component metrics

         arc   arc/e boolq hswag obkqa piqa  wino
DavidAU/Qwen3.5-27B-Claude-4.6-OS-INSTRUCT
mxfp8    0.675,0.827,0.900,0.750,0.496,0.800,0.721
qx86-hi  0.667,0.824,0.902,0.752,0.502,0.791,0.725

Qwen3.6-27B-Instruct
qx86-hi  0.637,0.798,0.911,0.775,0.442,0.807,0.737

Similar merge

Qwen3.6-27B-Qwopus-GLM-Instruct

         arc   arc/e boolq hswag obkqa piqa  wino
qx86-hi  0.656,0.826,0.910,0.776,0.474,0.812,0.739
qx64-hi  0.662,0.827,0.904

Quant    Perplexity      Peak Memory   Tokens/sec
qx86-hi  4.184 ± 0.027   32.36 GB      208
qx64-hi  4.184 ± 0.028   25.64 GB      216

Model recipe

models:
  - model: Qwen3.6-27B
    parameters:
      weight: 1.4
  - model: DavidAU/Qwen3.5-27B-Claude-4.6-OS-INSTRUCT
    parameters:
      weight: 0.6
merge_method: nuslerp
dtype: bfloat16
name: Qwen3.6-27B-Claude-4.6-OS

Detailed metrics coming soon.

-G

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("Qwen3.6-27B-Claude-4.6-OS-mxfp8-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)