forked from NVIDIA/cutlass
-
Notifications
You must be signed in to change notification settings - Fork 29
Pull requests: codeplaysoftware/cutlass-sycl
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Fix Variable Sequence Length Support for Flash Attention Decode
#362
opened May 7, 2025 by
muhammad-tanvir-1211
Loading…
Added the second release required FP16 fine-tuned GEMM kernels.
#361
opened May 7, 2025 by
tdeng5
Loading…
Fix variable length for Flash Attention prefill
#360
opened May 6, 2025 by
muhammad-tanvir-1211
Loading…
Fix Variable Sequence Length Support for Flash Attention Prefill + KV Cache
#359
opened May 6, 2025 by
muhammad-tanvir-1211
Loading…
Rename files and classes for Flash Attention
#354
opened May 2, 2025 by
muhammad-tanvir-1211
Loading…
Enable FP8_E5M2 GEMM and unify FP8 GEMM implementation with xe_mma.hpp
#352
opened May 2, 2025 by
sanchitintel
Loading…
[Please do not review] FP8 Grouped GEMM CollectiveMma
#351
opened May 1, 2025 by
sanchitintel
•
Draft
1 task
RFC: test out new syntax for launch with type deduction
#305
opened Apr 12, 2025 by
rolandschulz
Loading…
ProTip!
Mix and match filters to narrow down what you’re looking for.