Hugging Face's logo

waltgrace
/
mlx-expert-sniper

Image-Text-to-Text
MLX
English
apple-silicon
Mixture of Experts
mixture-of-experts
vision-language
gemma
falcon-perception
inference
Model card Files
xet
 Community
mlx-expert-sniper / src

Commit History

  1. Add Gemma 4-26B-A4B support: 4.15 tok/s on M4 Mac Mini⁠

    Nico +1: Claude Opus 4.6 (1M context)
    3f56a7b

  1. Add Gemma 4 MLX model class + preprocess

    waltgrace
    Verified signature
    cc1d5e2

  2. Add Gemma 4 MLX model class + preprocess

    waltgrace
    Verified signature
    4a30158

  3. Add Gemma 4 MLX model class + preprocess

    waltgrace
    Verified signature
    337dcd8

  4. Add Gemma 4 MLX model class + preprocess

    waltgrace
    Verified signature
    e2bb666

  5. Add Gemma 4 MLX model class + preprocess

    waltgrace
    Verified signature
    7409761

  6. Add Coder model + multi-model fixes

    waltgrace
    Verified signature
    46f93d5

  7. Add Coder model + multi-model fixes

    waltgrace
    Verified signature
    3a81687

  8. Add Coder model + multi-model fixes

    waltgrace
    Verified signature
    62e0ebd

  9. Add Coder model + multi-model fixes

    waltgrace
    Verified signature
    7878baa

  10. Add Coder model + multi-model fixes

    waltgrace
    Verified signature
    bded42e

  11. Add chat command + shared generate module

    waltgrace
    Verified signature
    3e350f6

  12. Add chat command + shared generate module

    waltgrace
    Verified signature
    8f6ad09

  13. Add 30B support + bias sweep results

    waltgrace
    Verified signature
    517be8a

  14. Add 30B support + bias sweep results

    waltgrace
    Verified signature
    8829376

  15. Add 30B support + bias sweep results

    waltgrace
    Verified signature
    22d3dc7

  16. Add Ollama-compatible serve command

    waltgrace
    Verified signature
    f7efda4

  17. Add Ollama-compatible serve command

    waltgrace
    Verified signature
    55fe21b

  18. Add download command: mlx-sniper download qwen3.5-35b

    waltgrace
    Verified signature
    e1498f9

  19. Add download command: mlx-sniper download qwen3.5-35b

    waltgrace
    Verified signature
    d694ca1

  1. Update src/mlx_expert_sniper/preprocess.py

    waltgrace
    Verified signature
    05209e2

  2. Update src/mlx_expert_sniper/expert_io.py

    waltgrace
    Verified signature
    bc2c2b1

  3. Update src/mlx_expert_sniper/engine.py

    waltgrace
    Verified signature
    7e0ed9b

  4. Update src/mlx_expert_sniper/coactivation.py

    waltgrace
    Verified signature
    482f969

  5. Update src/mlx_expert_sniper/calibrate.py

    waltgrace
    Verified signature
    9110344

  6. Update src/mlx_expert_sniper/cli.py

    waltgrace
    Verified signature
    1a7f86e

  7. Update src/mlx_expert_sniper/__init__.py

    waltgrace
    Verified signature
    0535634

  1. v0.2.0: Add Qwen3.5-35B-A3B support (5.78 tok/s, 19.5 GB on 16 GB RAM)

    waltgrace
    Verified signature
    d14a3c2

  1. v0.1.0: MoE expert sniping for MLX — run models larger than your RAM

    waltgrace
    Verified signature
    2d8a9bb