Commit History

  1. FP8 quantization: drop stale KD note, refresh PTQ metrics table⁠

    55c9d20

  1. Add FP8 vs fp16 evaluation results (1000 samples) to README

    411962a
  2. Update README: document FP8 quantization, remove usage/versions

    71f19ca
  3. Remove legacy pytorch_model.bin and modeling_gigaam.py from main

    a7f346b

  1. Adding `safetensors` variant of this model (#1)⁠

    vpermilp +1: SFconvertbot
    fbf69cd
  2. Duplicate from ai-sage/GigaAM-v3⁠

    vpermilp +1: Alexander4127
    7c307d9