Skip to content

Pull requests: pytorch/helion

New pull request New
126 Open 2,119 Closed
126 Open 2,119 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

Upgrade MI325 CI from ROCm 7.0 to 7.2 ciflow/rocm-mi300 CLA Signed This label is managed by the Meta Open Source bot. module: rocm
#1931 opened Apr 2, 2026 by v0i0 Contributor Loading…
1 task
2
[Auto-Recovery] Add crash recovery script for unrecoverable CUDA errors CLA Signed This label is managed by the Meta Open Source bot.
#1923 opened Apr 2, 2026 by yf225 Contributor Loading…
1
[Auto-Recovery] Add checkpoint tests CLA Signed This label is managed by the Meta Open Source bot.
#1920 opened Apr 2, 2026 by yf225 Contributor Loading…
1
Add HELION_AUTOTUNE_PRECOMPILE_REPS setting to run kernel multiple times during spawn precompile CLA Signed This label is managed by the Meta Open Source bot. fb-exported meta-exported
#1914 opened Apr 1, 2026 by yf225 Contributor Loading…
3
dbg CLA Signed This label is managed by the Meta Open Source bot.
#1911 opened Apr 1, 2026 by shunting314 Contributor • Draft
consolidate accuracy check APIs CLA Signed This label is managed by the Meta Open Source bot.
#1910 opened Apr 1, 2026 by shunting314 Contributor Loading…
8
[autotuner] Work around Triton two-dot shared-memory IMA on H100 CLA Signed This label is managed by the Meta Open Source bot.
#1905 opened Apr 1, 2026 by yf225 Contributor Loading…
3 tasks done
3
Add get_block_size_origin helper CLA Signed This label is managed by the Meta Open Source bot.
#1898 opened Mar 31, 2026 by jansel Contributor Loading…
[pallas-tpu] we are capping block sizes after they have been determined - that breaks in non-static shape regimes CLA Signed This label is managed by the Meta Open Source bot.
#1888 opened Mar 30, 2026 by v0i0 Contributor Loading…
1
[Pallas] Fix adjust_block_size_constraints to unblock test_bmm and test_attention_block_pointer CLA Signed This label is managed by the Meta Open Source bot.
#1881 opened Mar 30, 2026 by AmesingFlank Contributor Loading…
5
[metal] Auto-cap threadgroup size via CuteND/CuteFlattenedTileStrategy CLA Signed This label is managed by the Meta Open Source bot.
#1855 opened Mar 27, 2026 by aditvenk Contributor • Draft
3
no tensor descriptor for the 2d paralell gemm example CLA Signed This label is managed by the Meta Open Source bot.
#1848 opened Mar 27, 2026 by shunting314 Contributor Loading…
1
[Autotuner] Enable launch grid folding CLA Signed This label is managed by the Meta Open Source bot.
#1767 opened Mar 20, 2026 by bringlein Collaborator Loading…
38
[Autotuner] Add generation_ctx() context manager for each autotuner generation CLA Signed This label is managed by the Meta Open Source bot.
#1745 opened Mar 19, 2026 by yf225 Contributor Loading…
1 task done
4
[Experiment] LLM-guided autotuner MVP CLA Signed This label is managed by the Meta Open Source bot.
#1729 opened Mar 17, 2026 by kitrakrev • Draft
1
Fix #934: Add setting to disable 0/1 specialization CLA Signed This label is managed by the Meta Open Source bot.
#1678 opened Mar 13, 2026 by tianrengao Contributor • Draft
1
2
Fallback handling for streams without fileno and Rich LiveError in progress bar CLA Signed This label is managed by the Meta Open Source bot.
#1668 opened Mar 13, 2026 by Mistobaan Loading…
5
Fix error when operating on host values from torch.cuda.get_device_properties() CLA Signed This label is managed by the Meta Open Source bot.
#1623 opened Mar 6, 2026 by yf225 Contributor • Draft
1
3
Fix IndexError in Kernel.bind() when hl.specialize references a default argument CLA Signed This label is managed by the Meta Open Source bot.
#1622 opened Mar 6, 2026 by yf225 Contributor Loading…
2
Fill uninitialized memory with NaN during autotuner accuracy checks CLA Signed This label is managed by the Meta Open Source bot.
#1601 opened Mar 3, 2026 by mohaimingazi Loading…
4
Add triton.heuristics support to hl.triton_kernel() CLA Signed This label is managed by the Meta Open Source bot. language support
#1588 opened Mar 2, 2026 by mohaimingazi Loading…
8
Exclude block_ptr / tensor_descriptor from autotuner search space when index_dtype is int64 CLA Signed This label is managed by the Meta Open Source bot.
#1570 opened Feb 26, 2026 by yf225 Contributor Loading…
4
fix: rebenchmark crash during finishing CLA Signed This label is managed by the Meta Open Source bot.
#1523 opened Feb 20, 2026 by v0i0 Contributor Loading…
3
[WIP] Executable notebooksv2 CLA Signed This label is managed by the Meta Open Source bot.
#1496 opened Feb 18, 2026 by sekyondaMeta Contributor • Draft
[Helion + torch.compile] Support calling torch.compile on Helion kernel without eager warmup run CLA Signed This label is managed by the Meta Open Source bot.
#1456 opened Feb 13, 2026 by yf225 Contributor • Draft
8
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.