    # Qwen3.6-35B-A3B — EXL3

    [ExLlamaV3](https://github.com/turboderp-org/exllamav3) quantizations of
    [Qwen/Qwen3.6-35B-A3B](https://huggingface.co/Qwen/Qwen3.6-35B-A3B).

    Each bitrate lives on its own branch:

    | Branch | bpw | head bits |
    |---|---|---|
    | [4.00bpw](https://huggingface.co/MilesQLi/Qwen3.6-35B-A3B-exl3/tree/4.00bpw) | 4.00 | 6 |

| 4.50bpw | 4.50 | 6 | | 5.00bpw | 5.00 | 6 | | 6.00bpw | 6.00 | 6 | | 7.00bpw | 7.00 | 8 | | 8.00bpw | 8.00 | 8 |

    ```bash
    hf download MilesQLi/Qwen3.6-35B-A3B-exl3 --revision 4.00bpw --local-dir ./Qwen3.6-35B-A3B-exl3-4.00bpw
    ```

    Quantized with exllamav3 v1.4.6, default calibration (250 rows x 2048 tokens),
    `mul1` codebook. The vision tower and MTP layer are preserved.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

Image-Text-to-Text

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for MilesQLi/Qwen3.6-35B-A3B-exl3

Base model

Qwen/Qwen3.6-35B-A3B

Quantized

(782)

this model