Skip to content

Pull requests: RL-Align/RL-Kernel

New pull requestNew
42 Open 148 Closed
42 Open 148 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 reviewsReview requiredApproved reviewChanges requested
Assignee
Filter by who’s assigned
Assigned to nobodyLoading
Sort
Sort by
NewestOldestMost commentedLeast commentedRecently updatedLeast recently updatedBest match
Most reactions
👍👎😄🎉😕❤️🚀👀

Pull requests list

[PERF][distributed]: optimize deterministic ROCm collectives with HIP IPC platform: rocm Specific tasks specific to AMD graphics cards (such as CK, bpreshuffle/FA)
#357 opened Aug 29, 2026 by maxiaosong1124CollaboratorLoading…
1
[FEAT][distributed]: add deterministic ROCm/RCCL transport collectives platform: rocm Specific tasks specific to AMD graphics cards (such as CK, bpreshuffle/FA)
#356 opened Aug 28, 2026 by Flink-dddCollaboratorLoading…
1
feat(ascend): add deterministic collective Ascend C kernel Ascend
#355 opened Aug 28, 2026 by zhangj1anCollaboratorLoading…
1
bench(ffn): add H100 cuBLAS and vLLM comparisons platform: cuda Specific optimizations or bugs in NVIDIA graphics cards (such as FlashInfer, TMA optimizations)
#348 opened Aug 27, 2026 by frank-2077CollaboratorLoading…
1
[ROCm] Deterministic fused linear logp in Triton platform: rocm Specific tasks specific to AMD graphics cards (such as CK, bpreshuffle/FA)
#347 opened Aug 27, 2026 by KJLdefeatedCollaboratorLoading…
1
[Runtime][TP2 CP2] Publish strict bitwise runtime and user modes platform: cuda Specific optimizations or bugs in NVIDIA graphics cards (such as FlashInfer, TMA optimizations)
#344 opened Aug 26, 2026 by inaniloquenteeCollaboratorLoading…
1
feat(ascend): add prefix-shared attention Ascend C kernel Ascend
#340 opened Aug 25, 2026 by zhangj1anCollaboratorLoading…
1
feat(linear-logp): add batch-invariant fused SM90 operator
#337 opened Aug 24, 2026 by ThreeMonth03 • Draft
1
feat(ascend): add SwiGLU operator Ascend
#334 opened Aug 23, 2026 by erfgssLoading…
2
feat(logprob): add deterministic ROCm vocab-parallel path platform: rocm Specific tasks specific to AMD graphics cards (such as CK, bpreshuffle/FA)
#328 opened Aug 21, 2026 by hihaluemenContributorLoading…
1
feat(ffn): add deterministic distributed Triton FFN for ROCm platform: rocm Specific tasks specific to AMD graphics cards (such as CK, bpreshuffle/FA)
#325 opened Aug 20, 2026 by frank-2077CollaboratorLoading…
1
[WS2][gemm][2th merge] ffn collectives on test
#321 opened Aug 19, 2026 by frank-2077CollaboratorLoading…
1
[WS1][kernels] Deterministic attention Ascend C kernel Ascend
#320 opened Aug 19, 2026 by zhangj1anCollaboratorLoading…
2
feat(attention): add strict bitwise ROCm path platform: rocm Specific tasks specific to AMD graphics cards (such as CK, bpreshuffle/FA)
#319 opened Aug 19, 2026 by inaniloquenteeCollaboratorLoading…
1
[WS2][Logp] Expose the lse_vocab_tile_reduction ablation switch on linear-logp ops
#317 opened Aug 17, 2026 by zhangj1anCollaboratorLoading…
1
[WS2][Logp] Deterministic config option for operator platform: cuda Specific optimizations or bugs in NVIDIA graphics cards (such as FlashInfer, TMA optimizations)
#314 opened Aug 16, 2026 by KJLdefeatedCollaboratorLoading…
5
[CI][refator]: migrate GPU workflow needs-gpu-ci
#309 opened Aug 14, 2026 by Flink-dddCollaboratorLoading…
12
[WS2][Mismatch] Add Qwen3 FFN implementation factor
#308 opened Aug 13, 2026 by bitborneCollaboratorLoading…
19
[WS2][GEMM] Add Qwen3 FFN orchestration and consistency validation
#304 opened Aug 13, 2026 by bitborneCollaboratorLoading…
5 of 10 tasks
1
[WS2][PR5][Logp] Add deterministic DP-aware GRPO loss on the TP-aware logprob path
#291 opened Aug 12, 2026 by KJLdefeatedCollaboratorLoading…
10
[WS2][PR4][Logp] feat: add distributed TP/CP logprob drift report
#289 opened Aug 11, 2026 by hihaluemenContributorLoading…
9
[WS2] Ablation Matrix API v2.0 on top of PR230
#288 opened Aug 9, 2026 by zhangj1anCollaboratorLoading…
33
[WS2][PR3][Logp] Add deterministic vocab-parallel TP logprob reference
#265 opened Aug 5, 2026 by KJLdefeatedCollaboratorLoading…
6
docs(ws1): batch-invariance contract RFC (#101)
#264 opened Aug 5, 2026 by Billy1900Loading…
1
17
[WS2][PR2][Logp] feat: add TP=1 logprob comparison harness
#262 opened Aug 4, 2026 by hihaluemenContributorLoading…
11
ProTip! Updated in the last three days: updated:>2026-08-26.