- Notifications
You must be signed in to change notification settings - Fork38
Pull requests: pytorch/helion
Author
Uh oh!
There was an error while loading.Please reload this page.
Label
Uh oh!
There was an error while loading.Please reload this page.
Projects
Uh oh!
There was an error while loading.Please reload this page.
Milestones
Uh oh!
There was an error while loading.Please reload this page.
Reviews
Assignee
Assigned to nobodyLoading
Uh oh!
There was an error while loading.Please reload this page.
Sort
Pull requests list
Generalize examples with the DEVICE variable CLA SignedThis label is managed by the Meta Open Source bot.
#915 openedOct 13, 2025 byadam-smnkLoading…
[WIP] Adding new setting, autotune_effort=[none/quick/full] CLA SignedThis label is managed by the Meta Open Source bot.
#913 openedOct 12, 2025 bychoijon5Loading…
WIP: blackwell attn CLA SignedThis label is managed by the Meta Open Source bot.
#886 openedOct 10, 2025 byv0i0Loading…
internal way to set triton options CLA SignedThis label is managed by the Meta Open Source bot.
#880 openedOct 10, 2025 byv0i0Loading…
Catch missing cudnn error CLA SignedThis label is managed by the Meta Open Source bot.
#873 openedOct 10, 2025 byjanselLoading…
Add squeeze_and_excitation_net kernel CLA SignedThis label is managed by the Meta Open Source bot. fb-exported meta-exported
#870 openedOct 10, 2025 bymengluy0125Loading…
SupportThis label is managed by the Meta Open Source bot.
hl.arange() with non-power-of-2 input CLA Signed #862 openedOct 9, 2025 byyf225Loading…
Use helion.cdiv CLA SignedThis label is managed by the Meta Open Source bot.
#852 openedOct 8, 2025 byoulgenLoading…
Suggest use ofThis label is managed by the Meta Open Source bot.
@helion.kernel(index_dtype=torch.int64) if index offset is out of bound for int32 CLA SignedRemoveThis label is managed by the Meta Open Source bot.
triton_helpers.* usage in lifted device function arguments CLA SignedUpdate Discourse link of website to link to Helion channel on GPU mode discord CLA SignedThis label is managed by the Meta Open Source bot.
#844 openedOct 8, 2025 bychoijon5Loading…
Upgrade benchmarks to cuda13 CLA SignedThis label is managed by the Meta Open Source bot.
#826 openedOct 7, 2025 byoulgenLoading…
Upgrade to cuda 13 CLA SignedThis label is managed by the Meta Open Source bot.
#825 openedOct 7, 2025 byoulgenLoading…
[Benchmark] template attention kernel and test CLA SignedThis label is managed by the Meta Open Source bot.
#824 openedOct 7, 2025 bySibylauLoading…
add KL divergence backward helion kernel [attempt 2] CLA SignedThis label is managed by the Meta Open Source bot.
#805 openedOct 3, 2025 bywilliamwen42Loading…
[example] flex attention CLA SignedThis label is managed by the Meta Open Source bot.
#764 openedOct 1, 2025 byv0i0Loading…
[helion] backward support for swiglu CLA SignedThis label is managed by the Meta Open Source bot.
#756 openedOct 1, 2025 byshunting314Loading…
Add backward pass for softmax kernel CLA SignedThis label is managed by the Meta Open Source bot.
#744 openedOct 1, 2025 bykarthickaiLoading…
[RFC] Add support for device for loop indexing CLA SignedThis label is managed by the Meta Open Source bot.
[DO NOT MERGE]Make Puzzles Executable CLA SignedThis label is managed by the Meta Open Source bot.
#605 openedSep 17, 2025 bysekyondaMeta • Draft
[WIP] Add 2d and 3d indirect indexing support CLA SignedThis label is managed by the Meta Open Source bot.
ProTip! What’s not been updated in a month:updated:<2025-09-13.