COMMITS
June 1, 2026
R
Multiple updates and refactorings (#347)
Ray Wang committed
May 11, 2026
C
Update test_mega_moe.py
Chenggang Zhao committed
April 24, 2026
Z
Add various optimizations and Mega MoE benchmarks (#316)
Zhean Xu committed
April 17, 2026
C
[Public release 26/04] Introducing Mega MoE, FP4 Indexer and other features/fixes (#304)
Chenggang Zhao committed
March 22, 2026
R
Fix sync issue of TMEM alloc/dealloc (#292)
Ray Wang committed
February 25, 2026
X
February 3, 2026
R
Fix a sync issue in SM100 MQA logits (#285)
Ray Wang committed
January 16, 2026
Z
Multiple updates and refactorings (#280)
Zhean Xu committed
January 6, 2026
Z
Merge pull request #270 from yurekami/fix/sm90-archspec-bug
Zhean Xu committed
December 31, 2025
Y
fix: use SM90ArchSpec instead of SM100ArchSpec in sm90_bf16_k_grouped_gemm
yurekami committed
December 5, 2025
C
Update install.sh
Chenggang Zhao committed
A
Better error handling, code consistency, compile-time safety (#234)
AJ WISE committed
November 21, 2025
R
Multiple updates and refactorings (#231)
Ray Wang committed
November 19, 2025
Z
Fix sum_k * shape_m overflow
Zhean Xu committed
G
fix: prevent int32 overflow in k-grouped GEMM size calculations (#226)
Guoteng committed
R
Fix SM90 MQA logits (#229)
Ray Wang committed
November 14, 2025
R
Use larger MMA shape (#227)
Ray Wang committed
October 15, 2025
O
Merge pull request #220 from ko3n1g/ko3n1g/chore/revert-name-change
oliver könig committed
O
chore: Revert name change
oliver könig committed
O
chore: Rename project to ds-deem-gemm
oliver könig committed
O
Update publish.yml
oliver könig committed
O
Ko3n1g/chore/rename to deepgemm (#217)
oliver könig committed
October 14, 2025
O
ci: Fixes for pre-built wheels (#214)
oliver könig committed
October 11, 2025
C
Use CUDA runtime API to get device prop instead of ATen
Chenggang Zhao committed
October 10, 2025
O
chore: Build and store bdist wheels (#181)
oliver könig committed
October 9, 2025
J
Upgrade to CUTLASS 4.2.1 (#203)
Jun Jiang committed
October 1, 2025
P
Fix syntax errors and correct the conditional statements (#206)
PGFLMG committed
C
Fix version
Chenggang Zhao committed