daniel/h3-blackwell-runtime
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Tune Vortex NVFP4 scale geometry

Browse source
This commit is contained in:
Daniel Maddern 2026-08-15 02:15:19 +07:00
parent 1cc0ffdd00
commit 1fec77d2e9
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
src/h3_blackwell_runtime/nvfp4_quant.py
View file

@ -51,10 +51,10 @@ def _env_int(name: str, default: int) -> int:
def _vortex_scale_geometry(numel: int) -> tuple[int, int]:
threads = _env_int("H3_NVFP4_SCALE_THREADS", 256)
threads = _env_int("H3_NVFP4_SCALE_THREADS", 128)
if threads not in {128, 256, 512}:
raise ValueError(f"H3_NVFP4_SCALE_THREADS must be 128, 256, or 512, got {threads}")
default_blocks = min((numel + threads - 1) // threads, 4096)
default_blocks = min((numel + threads - 1) // threads, 256)
blocks = _env_int("H3_NVFP4_SCALE_BLOCKS", default_blocks)
if blocks <= 0:
raise ValueError(f"H3_NVFP4_SCALE_BLOCKS must be positive, got {blocks}")