Skip to content

Pull requests: pytorch/helion

New pull request New
131 Open 2,360 Closed
131 Open 2,360 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 UnicodeEncodeError in codegen under ASCII locale CLA Signed This label is managed by the Meta Open Source bot.
#2775 opened Jun 12, 2026 by karthickai Contributor • Draft
[cute] Paired fp8 decode (cvt.f16x2.e4m3x2) to close the GEMV perf gap CLA Signed This label is managed by the Meta Open Source bot.
#2774 opened Jun 12, 2026 by yushangdi Contributor • Draft
[cute] Fix warp-reduce hoist double-count for matmul split-K reductions CLA Signed This label is managed by the Meta Open Source bot.
#2773 opened Jun 12, 2026 by yushangdi Contributor • Draft
[cute] Vectorize fp8 loads in the SIMT matmul fallback CLA Signed This label is managed by the Meta Open Source bot.
#2772 opened Jun 12, 2026 by yushangdi Contributor • Draft
1
Lean kernel-artifact telemetry for cost-model dataset CLA Signed This label is managed by the Meta Open Source bot.
#2770 opened Jun 11, 2026 by IshanAryendu Contributor • Draft
5
[pallas] size emit_pipeline scratch from reshape-merged block-size products CLA Signed This label is managed by the Meta Open Source bot.
#2769 opened Jun 11, 2026 by choijon5 Contributor Loading…
[cute] Persist autotune winner from memory instead of recompiling CLA Signed This label is managed by the Meta Open Source bot.
#2768 opened Jun 11, 2026 by fulvius31 Collaborator Loading…
2
[examples] Add a simpler concat implementation CLA Signed This label is managed by the Meta Open Source bot.
#2766 opened Jun 11, 2026 by hinriksnaer Collaborator Loading…
[autotuner] Triton reduction seed heuristic (generalizable core) CLA Signed This label is managed by the Meta Open Source bot.
#2762 opened Jun 11, 2026 by calebmkim Contributor Loading…
[autotuner] Reduction fact layer: ReductionFact + AccumulatorFact + enriched MemoryOpFact CLA Signed This label is managed by the Meta Open Source bot.
#2761 opened Jun 11, 2026 by calebmkim Contributor Loading…
4
[examples] Edits to existing reduction example kernels for the seed-heuristic curriculum CLA Signed This label is managed by the Meta Open Source bot.
#2760 opened Jun 11, 2026 by calebmkim Contributor Loading…
2
Add a SymInt-free tensor specialization key for exact torch.Tensor args CLA Signed This label is managed by the Meta Open Source bot.
#2759 opened Jun 11, 2026 by yushangdi Contributor Loading…
[cute] Record measured-good B200 config for the fp8 scaled_mm example CLA Signed This label is managed by the Meta Open Source bot.
#2756 opened Jun 11, 2026 by yushangdi Contributor • Draft
[cute] Unified rolled TMA producer + hoisted K-loop predicates for tcgen05 CLA Signed This label is managed by the Meta Open Source bot.
#2755 opened Jun 11, 2026 by yushangdi Contributor • Draft
[cute] Unified rolled TMA producer + hoisted K-loop predicates for tcgen05 CLA Signed This label is managed by the Meta Open Source bot.
#2754 opened Jun 11, 2026 by yushangdi Contributor • Draft
Skip the measure("Kernel.bind") context manager when measurement is off CLA Signed This label is managed by the Meta Open Source bot.
#2752 opened Jun 11, 2026 by yushangdi Contributor • Draft
4
Move measure("Kernel.bind") off the cache-hit dispatch path CLA Signed This label is managed by the Meta Open Source bot.
#2751 opened Jun 11, 2026 by yushangdi Contributor • Draft
Collect kernel artifacts: device-IR node-link dump (.ir.jsonl) CLA Signed This label is managed by the Meta Open Source bot.
#2750 opened Jun 11, 2026 by IshanAryendu Contributor • Draft
Install a per-spec fast launcher that bypasses Triton's JITFunction.run CLA Signed This label is managed by the Meta Open Source bot.
#2749 opened Jun 10, 2026 by yushangdi Contributor Loading…
[Pallas] Add pallas_loop_type = 'outer_pipeline' CLA Signed This label is managed by the Meta Open Source bot.
#2744 opened Jun 10, 2026 by ethche Contributor • Draft
3
[Pallas] Fix attention example VMEM regression by making LSE 3D CLA Signed This label is managed by the Meta Open Source bot.
#2743 opened Jun 10, 2026 by norx1991 Contributor Loading…
Collect kernel artifacts and append-mode autotune telemetry with run_id CLA Signed This label is managed by the Meta Open Source bot.
#2737 opened Jun 10, 2026 by IshanAryendu Contributor • Draft
2
[Pallas] Rewrites of jagged reduction kernels in Pallas friendly ways. CLA Signed This label is managed by the Meta Open Source bot.
#2731 opened Jun 9, 2026 by thcmbs Collaborator Loading…
[Pallas] Test jagged carry with dynamic row counts CLA Signed This label is managed by the Meta Open Source bot.
#2722 opened Jun 8, 2026 by thcmbs Collaborator • Draft
[Autotuner] Support autotuing with non-dense mutated input CLA Signed This label is managed by the Meta Open Source bot.
#2721 opened Jun 8, 2026 by xiaohongchen1991 Contributor Loading…
ProTip! Type g i on any issue or pull request to go back to the issue listing page.