Commit History

Yesterday

  1. Quantize the LM head to NVFP4 (22 % faster decode, 2 GB smaller), add MTP and AWQ comparison to the card

    Verified signature
    659937b
  2. Card: image input via the plugin, vision check

    Verified signature
    b8e8307

2 days ago

  1. Add README.md

    Verified signature
    07c2b70