Qwen3.6-27B-Qwopus-GLM-bf16

This is a merge of the following models:

• Qwen/Qwen3.6-27B
• Jackrong/Qwopus3.5-27B-v3.5
• Jackrong/Qwen3.5-27B-GLM5.1-Distill-v1

         arc   arc/e boolq hswag obkqa piqa  wino
Qwen3.6-27B-Qwopus-GLM-Instruct
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

Component metrics

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

Qwen3.5-27B-GLM5.1-Distill-v1-Instruct
qx86-hi  0.619,0.775,0.900,0.735,0.440,0.801,0.713

Model recipe

models:
  - model: Jackrong/Qwopus3.5-27B-v3.5
    parameters:
      weight: 1.6
  - model: Jackrong/Qwen3.5-27B-GLM5.1-Distill-v1
    parameters:
      weight: 0.4
merge_method: nuslerp
dtype: bfloat16
name: Qwen3.5-27B-Qwopus3.5-GLM5.1

models:
  - model: Qwen/Qwen3.6-27B
    parameters:
      weight: 1.4
  - model: Qwen3.5-27B-Qwopus3.5-GLM5.1
    parameters:
      weight: 0.6
merge_method: nuslerp
dtype: bfloat16
name: Qwen3.6-27B-Qwopus3.5-GLM5.1-B

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("Qwen3.6-27B-Qwopus-GLM-bf16")

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)