Skip to content

[CUDA] Add opt-in direct scales for FP16 group-32 MatMulNBits M=1 - #33586

Open
Saiyam Bharara (saiyambharara) wants to merge 6 commits into
microsoft:mainfrom
saiyambharara:perf/32382-direct-scales
Open

Saiyam Bharara (saiyambharara) wants to merge 6 commits into
microsoft:mainfrom
saiyambharara:perf/32382-direct-scales

Commits

  1. Commits on Oct 25, 2026

  2. Commits on Oct 26, 2026

  3. Commits on Nov 2, 2026

  4. Commits on Nov 4, 2026

  5. Commits on Nov 9, 2026