Skip to content

Pull requests: ROCm/aiter

New pull request New
222 Open 2,271 Closed
222 Open 2,271 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

[TRITON] Fix test_moe_routing_sigmoid_top1_fused reference implementation tie-breaking bug Something isn't working ci:triton-355 triton
#2750 opened Apr 15, 2026 by brunomazzottiamd Contributor Loading…
1 task done
@brunomazzottiamd
1
Add tuned MoE configs for Qwen3-Next-80B-A3B FP8 on MI355X
#2748 opened Apr 15, 2026 by nholmber Loading…
1 task done
3
support a16wi4 for moe
#2747 opened Apr 15, 2026 by yadaish Contributor • Draft
1 task
1
fix(car): sglang prefill launch error kernel
#2745 opened Apr 15, 2026 by TennyWang1223 Contributor Loading…
1 task
1
Add glm5 70k 300 triton a8w8 blockscale configs
#2743 opened Apr 14, 2026 by amd-pedghazi Loading…
1 task done
7
Flydsl if else
#2740 opened Apr 14, 2026 by xudoyuan Contributor Loading…
1 task
7
fix moe splitk aot and jit ci:atom
#2738 opened Apr 14, 2026 by lalala-sh Contributor Loading…
1 task
1
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
fix gdr for vllm
#2736 opened Apr 14, 2026 by ganyi1996ppo Contributor • Draft
1 task
1
add tuner test suite
#2734 opened Apr 14, 2026 by yzhou103 Contributor Loading…
1 task
3
introduce g1u0 smoothquant int8 fused moe : fused_moe_gelu_sqi8
#2730 opened Apr 14, 2026 by tingqli Loading…
1 task
1
Add bf16 MLA decode kernel for gqa_ratio=64, qseqlen=1 (non-persistent)
#2729 opened Apr 14, 2026 by fangche123 Contributor Loading…
4
MI350 mla ps mode suppport nhead128,1 128,2 128,3 128,4 64,4 64,2 32,4 through kernel mla_a16w16_qh32_qseqlen4_gqaratio32_ps.co
#2727 opened Apr 14, 2026 by minmengdie Contributor Loading…
1 task
5
[FlyDSL MOE] update a8w4 moe
#2726 opened Apr 14, 2026 by Zzz9990 Contributor Loading…
1 task
32
flydsl implementation of a16w16 gemm
#2725 opened Apr 13, 2026 by omuhamma Contributor • Draft
1 task
@omuhamma
16
[TRITON] Add unified attention support to bench_models enhancement New feature or request triton
#2724 opened Apr 13, 2026 by lucas-santos-amd Contributor Loading…
1 task
@lucas-santos-amd
3
Ensure GemmTuner doesn't generate invalid input combinations for SplitK kernels
#2721 opened Apr 13, 2026 by apicciau Contributor • Draft
1 task
1
docs: comprehensive documentation overhaul
#2706 opened Apr 12, 2026 by sunway513 Collaborator Loading…
4 tasks
17
feat: add Gemma4 31B support (ProportionalRotaryEmbedding, rmsnorm dtype)
#2705 opened Apr 12, 2026 by ClementLinCF Collaborator Loading…
1 task done
5
Update quant.pyfix: add pack_dim to per_1x32_f4_quant for tl.dot_scaled RHS compatibility
#2704 opened Apr 12, 2026 by GeisYaO Loading…
2
fea(car): support custom group device
#2703 opened Apr 12, 2026 by TennyWang1223 Contributor Loading…
1 task
1
Add initial Windows support
#2699 opened Apr 11, 2026 by 0xDELUXA Contributor • Draft
1 task done
1
Add ROCm-versioned wheel naming to release workflow
#2698 opened Apr 11, 2026 by sunway513 Collaborator Loading…
3 tasks
7
Add FlyDSL fused RoPE + KV Cache backend
#2697 opened Apr 11, 2026 by coderfeli Collaborator Loading…
2 of 3 tasks
2
[Triton] Declare triton>=3.6.0 dependency ci:triton-355
#2695 opened Apr 10, 2026 by micmelesse Contributor Loading…
1 task
10
ProTip! Filter pull requests by the default branch with base:main.