SenseNova-Loop-9B-NVFP4-MLP / hf_quant_config.json

Commit History

Yesterday

  1. Quantize the LM head to NVFP4 (14 % faster decode, 1.5 GB smaller), add MTP and AWQ comparison to the card

    Verified signature
    ff00187

2 days ago

  1. Add hf_quant_config.json

    Verified signature
    1dcb0a2