Back to search
Top Contributors
763
commits
539
commits
450
commits
289
commits
263
commits
222
commits
180
commits
153
commits
131
commits
123
commits
Development repository for the Triton language and compiler
Stars
18.8k
Forks
2.7k
Watchers
18.8k
Open Issues
1.1k
Overall repository health assessment
No package.json found
This might not be a Node.js project
763
commits
539
commits
450
commits
289
commits
263
commits
222
commits
180
commits
153
commits
131
commits
123
commits
Revert "[language] Skip f16 to f32 promotion in max/min reductions" (#9921)
f97f66aView on GitHub[AMD] fix AsyncTDMCopyLocalToGlobalOp::verify bug about multi-cta (#9918)
8956d90View on GitHubrestrict prefetch pass to sm_80 target to fix performance regression (#9913)
01d2b7eView on GitHub[Standard] Skip f16 to f32 promotion in max/min reductions (#9903)
eab0c65View on GitHub[AMD][gfx1250] Pack f32 arith ops to use v_pk_* intrinsics (#9899)
d42e028View on GitHub[AMD] Generalize in-thread tree reduction to support ternary grouping for max/min (#9897)
bc79129View on GitHub[triton_kernels] Add other= to masked scale loads in _matmul.py to match _p_matmul.py (#9911)
7da7311View on GitHub[Gluon] Functional multi-CTA block scale MMA support (#9896)
798d24cView on GitHub[AMD] Make arith.select handling in canonicalize pointers more Robust (#8779)
4c69c79View on GitHub[PROTON][TEST] Add cupti graph replay heap growth repro (#9881)
e12b07aView on GitHub[AMD][Backend] Add OptimizeDescriptorEncoding pass for AMDGPU (#9792)
f50c8dfView on GitHub