Who contributed to PyTorch on July 20, 2026?

8 developers shipped this update, including Jean Schmidt, daisyden, cyy, RonitGandhi, orangeH25, calebmkim, jansel, and pytorchbot.

What were the notable PyTorch updates?

Fix cusparselt.cpp build when neither USE_CUSPARSELT nor USE_HIPSPARSELT is defined (#190496), [autotuner] promote pointwise seed to default on sm90/sm100, and [Test][Refactor] Decouple AotAutogradFallbackTests and HooksTests (#189373).

@pytorch

PyTorch and the broader machine-learning ecosystem

github ↗

Pick a date

Topics: Python AI / ML Full archive →

The Wire · Showcase

PYTORCH SHIPS BUILD FIX WHILE HELION TUNES H100 KERNELS TO DEFAULT

By RepoJournal · Filed 06:58 UTC on July 20, 2026 · About PyTorch

8 people shipped this

jansel @jansel 3 cited

Jean Schmidt @jeanschmidt 1 cited

daisyden @daisyden 1 cited

cyy @cyyever 1 cited

RonitGandhi @RonitGandhi 1 cited

calebmkim @calebmkim 1 cited

orangeH25 @orangeH25

pytorchbot @pytorchbot

A critical build regression in cusparselt.cpp is now patched, while the kernel autotuner promotes its pointwise seed heuristic to production defaults on H100 and B200.

In brief

Helion promoted pointwise seed to default on sm90/sm100. H100 autotuner now ships tuned kernels out of the box instead of burning first-run cycles. that's the real win here.
CuTe flash attention got a full tuning pass on dense ops. Helion's basically hand-tuning the GPU math PyTorch users rely on. fewer surprises at scale.
PyTorch fixed fast-mode gradcheck crashing on zero-element inputs. edge case? sure. but ops that fail silently on empty tensors are worse than ops that crash loud.
PyTorch fixed cusparselt.cpp build when neither sparse library is set. build stability isn't flashy but it's the difference between a repo that works and one that doesn't.

PyTorch's core build pipeline had a lurking fault: #178737 removed the file-level guard around cusparselt.cpp but left the function registration unprotected, causing compilation to fail on systems without cuSPARSELt or HIP support [1]. That's fixed. Meanwhile, Helion's autotuner is maturing its pointwise seed heuristic from experimental to default behavior on sm90 and sm100 architectures [2], a validation that the bandwidth-tile constants actually work at scale on H100 and B200. The move stays conservative elsewhere, keeping block_size=32 as the safe baseline for older hardware. On the attention front, Helion shipped a stacked PR series hardening its CuTe flash attention benchmark tooling [5] and adding FA4 runtime helpers [4], setting the stage for dense attention tuning [6]. Back in core, test infrastructure continues cleaning house: the S1/S2 decoupling strategy landed for AotAutogradFallbackTests [3], separating accelerator-agnostic tests from device-specific ones. Docs are purging decade-old Caffe2 references that no longer exist [7], and fast-mode gradcheck now handles zero-element inputs without crashing on forward AD [8].

Action items

→ Rebuild PyTorch on systems without cuSPARSELt or HIP to verify the cusparselt.cpp fix compiles cleanly pytorch/pytorch [plan]
→ If targeting H100 or B200, verify pointwise kernel defaults are now active by checking autotune-off behavior pytorch/helion [monitor]
→ Run gradcheck tests with zero-element tensors to ensure forward AD no longer crashes pytorch/pytorch [plan]

References

[1] Fix cusparselt.cpp build when neither USE_CUSPARSELT nor USE_HIPSPARSELT is defined (#190496) pytorch/pytorch
[2] [autotuner] promote pointwise seed to default on sm90/sm100 ↗ pytorch/helion
[3] [Test][Refactor] Decouple AotAutogradFallbackTests and HooksTests (#189373) pytorch/pytorch
[4] [cutedsl] Add FA4 flash runtime helpers ↗ pytorch/helion
[5] [cutedsl] Harden attention benchmark tooling ↗ pytorch/helion
[6] [cutedsl] Tune dense CuTe flash attention ↗ pytorch/helion
[7] Remove stale Caffe2 references from docs (#190430) pytorch/pytorch
[8] Fix fast-mode gradcheck forward AD crash for zero-element inputs (#190437) pytorch/pytorch

Quick answers

What shipped in PyTorch on July 20, 2026?: A critical build regression in cusparselt.cpp is now patched, while the kernel autotuner promotes its pointwise seed heuristic to production defaults on H100 and B200. In total, 32 commits and 4 pull requests landed.
Who contributed to PyTorch on July 20, 2026?: 8 developers shipped this update, including Jean Schmidt, daisyden, cyy, RonitGandhi, orangeH25, calebmkim, jansel, and pytorchbot.
What were the notable PyTorch updates?: Fix cusparselt.cpp build when neither USE_CUSPARSELT nor USE_HIPSPARSELT is defined (#190496), [autotuner] promote pointwise seed to default on sm90/sm100, and [Test][Refactor] Decouple AotAutogradFallbackTests and HooksTests (#189373).

HELION BACKEND REFACTOR UNLOCKS MULTI-DEVICE COMPILATION; DYNAMO FIXES FORWARD-AD TANGENT BUG

Helion shipped a four-part backend architecture overhaul that finally separates printer, lowering, and backend registration logic by device, eliminating the need to edit 17 core files for every new backend.

python 51 shipped 2-min read

@pytorch 3 days ago

HELION SHIPS B200 AUTOTUNER, PYTORCH PLUGS CUDA DUPLICATION, TORCHRL ADDS PULL-BASED MONITORING

Helion's stacked reduction-seed redesign targets NVIDIA's B200 with dedicated SM100 heuristics, while PyTorch core eliminates duplicate CUDA wrappers in NCCL2 and TorchRL lands its first operational monitoring layer.

python 82 shipped 1-min read

@pytorch 5 days ago

EXECUTORCH SLASHES CONV LATENCY, PYTORCH CORE FIXES FLEXATTENTION SPECIALIZATION BUG

ExecutorTorch's ARM backend just folded scalar multiplication into convolution weights, eliminating post-conv ops entirely, while PyTorch core patches a FlexAttention bug that was over-specializing dynamic sequence lengths in block masks.

python 69 shipped 1-min read

@pytorch 7 days ago

INDUCTOR FIXES NESTED REDUCTION BLOCK POINTER BUG, TORCHRL UNIFIES CHECKPOINTING

Inductor's Triton codegen was silently corrupting fused broadcasts and reductions by losing track of block pointer advancement across outer loops, now patched [ref:3].

python 77 shipped 1-min read

Elsewhere on the wire

@cachyos about 18 hours ago

LINUX 7.2-RC4 LANDS WITH GAMING SCHEDULER, ANANICY OVERHAULS PROCESS TUNING

CachyOS shipped kernel 7.2-rc4 with a new gaming-optimized scheduler branch and AMD display fixes, while ananicy-rules yanked normal scheduling from most process types in a major behavioral shift.

infra 56 shipped 1-min read

Elixir & Phoenix about 18 hours ago

ELIXIR COMPILER GETS 20% FASTER TYPE CHECKING

Multi-clause function analysis now runs significantly faster after the core team eliminated unnecessary list allocations in disjoint argument checking.

elixir 6 shipped 1-min read

@dotnet about 18 hours ago

MACIOS FIXES SYMBOL UPLOAD GAP, MSBUILD PULLS ARCADE UPDATES

Apple's App Store Connect will finally see proper crash symbolication for .NET iOS apps after macios shipped a fix that embeds symbol files directly in the .ipa.

dotnet 21 shipped 1-min read

FastAPI & Pydantic about 18 hours ago

PYDANTIC DOCS GET GRAMMAR CLEANUP

Pydantic merged a four-typo fix across docs and skill definitions that slipped past CI.

python 2 shipped 1-min read

Want every project, not just this one?

Follow @pytorch