Skip to content

Pull requests: ROCm/aiter

New pull request New
102 Open 919 Closed
102 Open 919 Closed
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
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

fix the mha pytest err
#1146 opened Oct 9, 2025 by minmengdie Loading…
1 task
1
Fix sgl deepseek error with cudagraph plus torch.compile
#1145 opened Oct 9, 2025 by ZhangLirong-amd Loading…
1 task
[MI35X] Fix FA bwd result mismatch when sq==1
#1144 opened Oct 9, 2025 by slippedJim Loading…
1 task
CI: Add MI35X runners for Aiter CI
#1141 opened Oct 9, 2025 by gyohuangxin Loading…
4
add asm pa qlen4 for gqa 10
#1140 opened Oct 9, 2025 by junhaha666 Loading…
1 task
1
fix the error in the fwd v3 api when the group mode not supported swa yet
#1139 opened Oct 9, 2025 by minmengdie Loading…
1 task
1
Make soltype default is None
#1138 opened Oct 9, 2025 by ZhangLirong-amd Loading…
1 task
[BugFix] Change Vskip Selection Logic
#1136 opened Oct 8, 2025 by vllmellm Loading…
1 task
1
1
fmha v3 API only generate for supported targeting GPU arch
#1134 opened Oct 7, 2025 by HollowMan6 Loading…
1 task done
16
Various benchmark script fixes
#1133 opened Oct 7, 2025 by jayfurmanek • Draft
Bump composable_kernel for fmha 192x128 head dims commit
#1130 opened Oct 6, 2025 by samremes • Draft
1 task
[CK_TILE] Temporarily disable k length=1 test cases in seqence padding
#1129 opened Oct 3, 2025 by Jeff-Huang Loading…
Add openpyxl to optional deps
#1128 opened Oct 2, 2025 by amd-asalykov Loading…
1 task
3
[Triton] e2e fused MoE for small N and fp8 blockscale MoE benching
#1126 opened Oct 2, 2025 by juuso-oskari Loading…
@juuso-oskari @Chi-Chu319
5
update sparse mla kernel
#1125 opened Oct 2, 2025 by Zzz9990 • Draft
1 task
lean_attention: add GQA support across kernel and wrapper; add tests
#1123 opened Oct 1, 2025 by kesavanramakrishnan Loading…
3
[Triton] FP4 GEMM weight shuffle support and tunning
#1120 opened Sep 30, 2025 by k50112113 Loading…
integrate cktile moe into fused_moe interface
#1111 opened Sep 28, 2025 by zhiding512 Loading…
Mla splitkv enhance split alg inte
#1108 opened Sep 28, 2025 by fangche123 Loading…
1 task
3
Tuning Utility update
#1102 opened Sep 27, 2025 by valarLip Loading…
1 task
5
[MI300X] tune Gemm performance for llama 70b/405b
#1094 opened Sep 26, 2025 by zhuyuhua-v Loading…
add module_gemm_a8w8_blockscale in aot_build list
#1091 opened Sep 26, 2025 by ZJLi2013 Loading…
Sixifang/hip pa
#1089 opened Sep 26, 2025 by valarLip Loading…
1 task
4
CI: Debug pybind11 dep in vLLM benchmark
#1088 opened Sep 26, 2025 by gyohuangxin • Draft
[config] tune gemm and moe for qwen3 480b ptpc model on MI308
#1084 opened Sep 25, 2025 by gbyu-amd Loading…
5
ProTip! Exclude everything labeled bug with -label:bug.