AlphaOxO/Param2-17B-A2.4B-Thinking-4bits-mlx

Param2-17B-A2.4B-Thinking — MLX 4bit

Quantized to 4 bits per weight.

Converted from bharatgenai/Param2-17B-A2.4B-Thinking using mlx-lm.

Requirements

The param2moe model type is support in my mlx-lm fork(AlphaKure/mlx-lm.

Until test finish. I will send a PR to offical repo.

Update at: 2026/10/13

pip install git+https://github.com/AlphaKure/mlx-lm.git@88bb9fb88fe233aa21776fe76306c459b80453b2

Use

mlx_lm.generate --model AlphaOxO/Param2-17B-A2.4B-Thinking-4bits-mlx --prompt "hello"

from mlx_lm import load, generate

model, tokenizer = load("AlphaOxO/Param2-17B-A2.4B-Thinking-4bits-mlx")
prompt = tokenizer.apply_chat_template(
    [{"role": "user", "content": "hello"}], add_generation_prompt=True
)
print(generate(model, tokenizer, prompt=prompt))