Skip to content

Pull requests: ROCm/aiter

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

Add tuned a8w8 blockscale GEMM config for Qwen3-Next-80B-A3B on MI355X
#2868 opened Apr 22, 2026 by nholmber Contributor Loading…
1 task done
Fix Correction Issues in batched_gemm_a8w8 for ASYNC_COPY Enabled bug Something isn't working ci:triton-355 triton
#2867 opened Apr 22, 2026 by nidal567 Contributor Loading…
1 task done
Add qwen3.5 397b mxfp4 fmoe tuning
#2865 opened Apr 22, 2026 by mqhc2020 Contributor Draft
1 task
kimi a16wi4 moe support
#2863 opened Apr 22, 2026 by yadaish Contributor Loading…
1 task
Bump CK for a stride fix in CKTile Block-Scale GEMM ci:all
#2862 opened Apr 22, 2026 by samremes Contributor Draft
1 task
update qwen3next config
#2861 opened Apr 22, 2026 by ganyi1996ppo Contributor Loading…
1 task
Add torch in gemm a16w16 tune
#2860 opened Apr 22, 2026 by yzhou103 Contributor Loading…
1 task
docs: polish README with ecosystem, news, and performance highlights
#2859 opened Apr 22, 2026 by carlushuang Collaborator Loading…
Dev/aot fix
#2856 opened Apr 22, 2026 by zhiding512 Contributor Loading…
[FLYDSL] Add GDR prefill k5 kernel
#2854 opened Apr 22, 2026 by huizzhan Contributor Draft
1 task
make rmsnorm quant fusion support gemma
#2853 opened Apr 22, 2026 by ganyi1996ppo Contributor Loading…
1 task
MLA PS mode support nhead8,2 in MI308
#2852 opened Apr 22, 2026 by minmengdie Contributor Loading…
1 task
Update kimik2 FP4 tuned fMoE config
#2845 opened Apr 21, 2026 by okorzh-amd Loading…
1 task done
fav3 kernel with improved softmax
#2830 opened Apr 21, 2026 by antsaukk Contributor Draft
1 task
Fix sliding window mtp
#2829 opened Apr 21, 2026 by fsx950223 Contributor Loading…
1 of 2 tasks
Fused AR + RMSNorm + per-group FP8 quant: optional bf16 side-output
#2823 opened Apr 20, 2026 by hubertlu-tw Contributor Loading…
3 tasks
Flydsl implementation of a8w8 blockscale for gfx1250 (WIP)
#2818 opened Apr 20, 2026 by omuhamma Contributor Draft
1 task
ProTip! What’s not been updated in a month: updated:<2026-03-22.