-
Notifications
You must be signed in to change notification settings - Fork 78
Pull requests: Tencent/hpc-ops
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Add independent K scale support in FP8 prefill attention
#40
opened Apr 3, 2026 by
xueyangcs
Loading…
decode bf16 smallm: support arbitrary 1<heads_per_group<=8 via direct Q/Y GMEM when TMA unsuitable
#37
opened Apr 2, 2026 by
Religious-J
Loading…
ProTip!
Mix and match filters to narrow down what you’re looking for.