-
Notifications
You must be signed in to change notification settings - Fork 485
Pull requests: ROCm/aiter
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[JIT][Bugfix] Make blob codegen cache publication transactional
#4800
opened Aug 17, 2026 by
Ruye-aa
Loading…
config(glm5): retune MXFP4 MoE token=8 to a4w4
#4798
opened Aug 17, 2026 by
jiaryang
Loading…
4 tasks done
fix: generate CK FMHA kernels for gfx1250
#4797
opened Aug 17, 2026 by
tomjen12
Contributor
Loading…
[module_mla_reduce] detorch reduce.cu + delete dead no_redundant Python port
ci:atom
#4796
opened Aug 17, 2026 by
amd-ruitang3
Contributor
Loading…
ci: add network diagnostics for test artifact downloads
#4795
opened Aug 17, 2026 by
gyohuangxin
Member
Loading…
Resolve the pybind develop-path lookups once instead of per call
#4793
opened Aug 17, 2026 by
fangche123
Contributor
Loading…
1 task
fix(tuned_gemm): gate skinny GEMM default to archs with a real kernel
#4791
opened Aug 17, 2026 by
yichiche
Contributor
Loading…
[WIP]update gemm stage2 v2 kernel
#4789
opened Aug 17, 2026 by
charlieguo1106
Contributor
•
Draft
1 task
fix(custom_all_reduce): use SYSTEM scope + ACQUIRE ordering for cross-device signal loads
#4786
opened Aug 16, 2026 by
hekhong-png
Loading…
[feat] mega-moe stage2 for gfx1250
#4785
opened Aug 16, 2026 by
yanboshao
Contributor
Loading…
1 task
[gfx950][FlyDSL] Add direct dense A4W4 MXFP4 GEMM
#4782
opened Aug 16, 2026 by
LiuYinfeng01
Contributor
•
Draft
6 tasks done
[gfx950] Retune the small-M tiles and wave counts in the A16W16 fallback config
#4781
opened Aug 16, 2026 by
akii96
Contributor
Loading…
[Bugfix] Handle GroupNorm autocast safely
#4779
opened Aug 15, 2026 by
akshatvishu
Loading…
5 of 8 tasks
[gfx1100] Enable RDNA3 in arch allow-list + Triton GEMM A8W8 tuning config
#4778
opened Aug 15, 2026 by
okone1995
Loading…
perf(mla): chunk the non-FP4 gather_kv_b_proj over KV
#4776
opened Aug 15, 2026 by
zejunchen-zejun
Contributor
Loading…
[Attention] Expose paged MQA SplitKV override
#4775
opened Aug 15, 2026 by
AMD-yanfeiwang
Contributor
•
Draft
[gfx950][FlyDSL] Add dense BF16 x MXFP4 GEMM
#4772
opened Aug 15, 2026 by
LiuYinfeng01
Contributor
Loading…
5 tasks done
[Triton][FMHA] Fused paged-prefill kernel for page_size=1, head_dim 64/256
#4771
opened Aug 15, 2026 by
amd-ethany
Loading…
1 task done
[Bugfix] Include headers necessary for ROCm 10.0.0
#4768
opened Aug 14, 2026 by
rjrock
Loading…
1 task done
feat(moe): consume prepared stage1 activation scales
#4762
opened Aug 14, 2026 by
JohnQinAMD
Contributor
Loading…
Previous Next
ProTip!
Exclude everything labeled
bug with -label:bug.