Skip to content

Pull requests: pytorch/helion

New pull request New
33 Open 998 Closed
33 Open 998 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

Clean up distributed examples path refs CLA Signed This label is managed by the Meta Open Source bot.
#1241 opened Dec 10, 2025 by yf225 Loading…
fix for circular dependency CLA Signed This label is managed by the Meta Open Source bot. fb-exported meta-exported
#1236 opened Dec 9, 2025 by mengluy0125 Loading…
1
[WIP][xpu][fix] improve XPU host time in tuning by creating less events CLA Signed This label is managed by the Meta Open Source bot.
#1230 opened Dec 8, 2025 by jianyizh • Draft
Allow using hl.specialize to specialize on tensor strides CLA Signed This label is managed by the Meta Open Source bot.
#1215 opened Dec 5, 2025 by yf225 Loading…
8
[wip][Interpret Mode] Better error message for broadcast shape mismatch CLA Signed This label is managed by the Meta Open Source bot.
#1211 opened Dec 5, 2025 by yf225 • Draft
1
Use torch._dynamo.mark_static() API to allow tensor shape specialization outside of the kernel code CLA Signed This label is managed by the Meta Open Source bot.
#1210 opened Dec 5, 2025 by yf225 Loading…
5
[xpu][fix] Change min_dot_size call due to Triton API Change CLA Signed This label is managed by the Meta Open Source bot.
#1173 opened Nov 28, 2025 by Stonepia Loading…
1
[wip] Support hl.barrier for mega-kernels CLA Signed This label is managed by the Meta Open Source bot.
#1151 opened Nov 20, 2025 by jansel • Draft
Add OpenEvolve-based Autotuner for Helion GPU Kernels CLA Signed This label is managed by the Meta Open Source bot.
#1082 opened Nov 4, 2025 by mycpuorg • Draft
1
Added dynamic-shape 0/1 bucketing: "zero_nonzero" env var CLA Signed This label is managed by the Meta Open Source bot.
#1053 opened Oct 30, 2025 by Itssshikhar Loading…
1
@yf225
27
Use matmul fwd direclty in autograd for performance CLA Signed This label is managed by the Meta Open Source bot.
#1045 opened Oct 28, 2025 by tianrengao Loading…
2
Disallow both interpret modes active CLA Signed This label is managed by the Meta Open Source bot.
#1023 opened Oct 25, 2025 by yf225 • Draft
blackwell attn with acc and stock triton support CLA Signed This label is managed by the Meta Open Source bot.
#996 opened Oct 20, 2025 by v0i0 • Draft
Add simplified se_block kernel CLA Signed This label is managed by the Meta Open Source bot. fb-exported meta-exported
#989 opened Oct 20, 2025 by mengluy0125 Loading…
6
scripts for blackwell attn measurement CLA Signed This label is managed by the Meta Open Source bot.
#977 opened Oct 16, 2025 by v0i0 Loading…
Add epilogue subtiling CLA Signed This label is managed by the Meta Open Source bot.
#948 opened Oct 15, 2025 by PaulZhang12 Loading…
26
Validate user provided settings CLA Signed This label is managed by the Meta Open Source bot.
#947 opened Oct 15, 2025 by choijon5 Loading…
3
Use helion.cdiv CLA Signed This label is managed by the Meta Open Source bot.
#852 opened Oct 8, 2025 by oulgen Loading…
Upgrade benchmarks to cuda13 CLA Signed This label is managed by the Meta Open Source bot.
#826 opened Oct 7, 2025 by oulgen Loading…
Upgrade to cuda 13 CLA Signed This label is managed by the Meta Open Source bot.
#825 opened Oct 7, 2025 by oulgen Loading…
1
[Benchmark] template attention kernel and test CLA Signed This label is managed by the Meta Open Source bot.
#824 opened Oct 7, 2025 by Sibylau Loading…
3
add KL divergence backward helion kernel [attempt 2] CLA Signed This label is managed by the Meta Open Source bot.
#805 opened Oct 3, 2025 by williamwen42 Loading…
2
Add advanced compiler configurations CLA Signed This label is managed by the Meta Open Source bot.
#793 opened Oct 3, 2025 by jansel • Draft
8
[example] flex attention CLA Signed This label is managed by the Meta Open Source bot.
#764 opened Oct 1, 2025 by v0i0 Loading…
14
[RFC] Add support for device for loop indexing CLA Signed This label is managed by the Meta Open Source bot.
#673 opened Sep 24, 2025 by oulgen • Draft
1
ProTip! Mix and match filters to narrow down what you’re looking for.