HN Explorer

opcode84

↗ HN profile [ 47.7 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Nvidia trains 10T model in 4 bit precision (NVFP4) (developer.nvidia.com)
    9 points 3
  2. Nvidia announces 4-bit training with NVFP4 (developer.nvidia.com)
    2 points 1
  3. Blackwell Brings Native Support for NVFP4: 4-Bit Inference with High Accuracy (developer.nvidia.com)
    1 points 1
  4. New microscaling data formats supported in Blackwell and Triton (triton-lang.org)
    3 points 1
  5. NVIDIA confirms Blackwell support for OCP microscaling formats MXFP8, MXFP4 (developer.nvidia.com)
    4 points 1
  6. Standardizing next-generation narrow precision data formats for AI (opencompute.org)
    103 points 52