RynnBrain-8B-qx86-hi-mlx

Brainwaves

Alibaba-DAMO-Academy/RynnBrain-8B
qx86-hi  0.617,0.851,0.886,0.751,0.468,0.810,0.730

Alibaba-DAMO-Academy/RynnBrain-30B-A3B
qx86-hi  0.641,0.859,0.895,0.786,0.474,0.823,0.751

Base models

Qwen3-VL-8B-Instruct
qx86-hi  0.455,0.596,0.872,0.543,0.424,0.736,0.593

Qwen3-VL-30B-A3B-Instruct
qx86-hi  0.439,0.541,0.894,0.619,0.430,0.764,0.592

Use with mlx

pip install mlx-lm

from mlx_lm import load, generate

model, tokenizer = load("RynnBrain-8B-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)