mlx-community

/

Qwen3.5-397B-A17B-4bit

like 5

Follow

MLX Community 8.82k

Yes, text-only; Q4.

I can't use this version via lm studio, say's: Error when loading model: ValueError: Model type qwen3_5_moe not supported. Is this normal?

i got this error in vlmx too? Do you have update ?

<frozen runpy>:128: RuntimeWarning: 'mlx_vlm.generate' found in sys.modules after import of package 'mlx_vlm', but prior to execution of 'mlx_vlm.generate'; this may result in unpredictable behaviour
Calling `python -m mlx_vlm.generate ...` directly is deprecated. Use `mlx_vlm generate` or `python -m mlx_vlm generate` instead.
Fetching 56 files: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 56/56 [00:00<00:00, 231638.09it/s]
Download complete: : 0.00B [00:00, ?B/s]                                                                                                                    | 0/56 [00:00<?, ?it/s]
==========
Files: ['L.jpg'] 

Prompt: <|im_start|>user
<|vision_start|><|image_pad|><|vision_end|>Describe this image.<|im_end|>
<|im_start|>assistant
<think>

The user wants a description of the provided image.

1.  **Identify the object:** It's a Blu-ray disc case.
2.  **Identify the main subject:** The cover features a close-up of Jackie Chan's face.
3.  **Identify the text:**
    *   Top: "JACKIE CHAN'S FIRST STRIKE" in large, bold, gold/white letters.
    *   Left spine: Chinese characters (vertical)
==========
Prompt: 170 tokens, 220.026 tokens-per-sec
Generation: 100 tokens, 37.009 tokens-per-sec
Peak memory: 224.582 GB