Skip to content

Pull requests: ROCm/aiter

New pull request New
228 Open 2,296 Closed
228 Open 2,296 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

gemma quant
#2789 opened Apr 19, 2026 by ganyi1996ppo Contributor Loading…
1 task
5
Dev/a16wi4 a8w4 merge
#2786 opened Apr 18, 2026 by yadaish Contributor • Draft
1 task
32
Fix LDS Exhaustion in fused_gemm_afp4wfp4_a16w16 for ASYNC_COPY Enabled bug Something isn't working ci:triton-355 triton
#2784 opened Apr 17, 2026 by nidal567 Contributor Loading…
1 task done
1
Gluon gemma8w8 blockscale wrap-up
#2783 opened Apr 17, 2026 by amirumoAMD Contributor • Draft
1 task
1
[FlyDSL MOE] fix _fq suffix compatibility in kernel name parsing
#2782 opened Apr 17, 2026 by okorzh-amd Loading…
1 task
@okorzh-amd
2
Add mutates_args=[] to gemm_a4w4 torch_compile_guard to fix torch.compile crash (fix for #2780)
#2781 opened Apr 17, 2026 by ColinZ22 Loading…
1 task done
4
[Don't merge] Gluon pa bad case reproducer
#2779 opened Apr 17, 2026 by ganyi1996ppo Contributor • Draft
1 task
2
[attention] refactor hip kl
#2778 opened Apr 17, 2026 by amd-ruitang3 Contributor Loading…
1 task
11
[FLYDSL]: aiter flydsl if const_expr
#2776 opened Apr 17, 2026 by xudoyuan Contributor • Draft
1 task
1
Xiaobing/gptoss small m
#2775 opened Apr 17, 2026 by XiaobingSuper Contributor • Draft
1 task
1
[HIP] add chunk_gated_delta_rule_fwd_h_hip kernel for prefill GDN support
#2774 opened Apr 17, 2026 by yiijin Loading…
1 task
12
[DO NOT MERGE] CI: install Triton wheel before Aiter standard tests
#2773 opened Apr 17, 2026 by gyohuangxin Member Loading…
2 of 3 tasks
6
make cache op support non-contiguous num_blocks dim
#2772 opened Apr 17, 2026 by ganyi1996ppo Contributor Loading…
1 task
5
[Triton]: Add MoE a16w4
#2770 opened Apr 16, 2026 by rahulbatra85 Contributor • Draft
1 task
4
docs(skills): add AITER Claude/Cursor skill set with validator
#2769 opened Apr 16, 2026 by ChuanLi1101 Loading…
4 tasks done
2
Add SGLang/vLLM/ATOM integration tests to nightly workflow
#2767 opened Apr 16, 2026 by kiran-thumma Collaborator • Draft
5 tasks
1
Satya.gfx12 mxfp4 gemm scale buffer load
#2765 opened Apr 16, 2026 by Boss2002n Contributor • Draft
1 task
[Triton] Fix Flash Attention Cuda Graphs issues
#2764 opened Apr 16, 2026 by micmelesse Contributor • Draft
1 task
1
feat(moe): support multi-B weight tensors (DWDP) in FlyDSL MOE kernels
#2762 opened Apr 16, 2026 by AMD-yanfeiwang Contributor • Draft
1 task
1
[DO NOT MERGE] CI: try downloading pre-built Triton wheel before building from source
#2761 opened Apr 16, 2026 by mengfei-jiang Loading…
1 task done
1
[ROPE] refactor hip kls
#2754 opened Apr 16, 2026 by amd-ruitang3 Contributor Loading…
1 task
8
Add tuned MoE configs for Qwen3-Next-80B-A3B FP8 on MI355X
#2748 opened Apr 15, 2026 by nholmber Contributor Loading…
1 task done
3
support a16wi4 for moe
#2747 opened Apr 15, 2026 by yadaish Contributor • Draft
1 task
1
Add glm5 70k 300 triton a8w8 blockscale configs
#2743 opened Apr 14, 2026 by amd-pedghazi Loading…
1 task done
7
Revert max_size from 1GB to 128MB to fix KV cache regression
#2737 opened Apr 14, 2026 by AMD-yanfeiwang Contributor • Draft
1 task
1
ProTip! What’s not been updated in a month: updated:<2026-03-19.