PyTorch 2.14.0 Release

PyTorch 2.14.0 Release pytorch / pytorch Public Notifications You must be signed in to change notification settings Fork 30.5k Star 103k PyTorch 2.14.0 Release Latest Latest ethche released this 02 Sep 17:40 · 2597 commits to main since this release v2.14.0 2b3ec34 This commit was created on GitHub.com and signed with GitHub’s verified signature . GPG key ID: B5690EEEBB952194 Verified Learn about vigilant mode .

This HardwareLaunch is relevant to the technology intelligence record because it involves Apple, Meta, Intel, GitHub. The source article should remain the factual reference for follow-up coverage.

• pytorch / pytorch Public Notifications You must be signed in to change notification settings Fork 30.5k Star 103k PyTorch 2.14.0 Release Latest Latest ethche released this 02 Sep 17:40 · 2597 commits to main since this release v2.14.0 2b3ec34 This commit was created on GitHub.com and signed with GitHub’s verified signature .
• GPG key ID: B5690EEEBB952194 Verified Learn about vigilant mode .
• PyTorch 2.14.0 Release Notes Highlights Backwards Incompatible Changes Deprecations New Features Improvements Bug fixes Performance Documentation Security Developers Highlights NVGEMM brings CuTeDSL-generated CUTLASS kernels to Inductor, with epilogue fusion, scaled and NVFP4 GEMM, and grouped-reduction epilogues autotuned alongside Triton and ATen torch.switch generalizes torch.cond to multi-way branching, and torch.while_loop can now be captured in a CUDA graph Declarative dynamic shapes via @dynamic_spec , shared across torch.compile , torch.export and make_fx Experimental torch.compile support for complex-valued tensors : Opt-in support decomposes supported complex operations into real and imaginary computations, enabling compiler backends to optimize more complex-number workloads.
• A preview of our rewritten NCCL backend for PyTorch , ported from torchcomms, implementing the full collective contract with nonblocking communicators and eager communicator splitting and advanced features such as fault tolerance and windows designed as a drop-in replacement of existing NCCL c10d backend Fault tolerance becomes a first-class c10d concept , with in-place process-group reconfiguration, one-sided RMA windows, and a Flight Recorder that works for any backend rather than only NCCL Apple Silicon gains native linear algebra , including Jacobi-kernel SVD, eigh , QR and Cholesky, alongside a five-part reduction rewrite and a further MPSGraph to Metal kernel migration Broader platform support : ROCm 7.14 wheels are produced from the TheRock pip SDK, Intel XPU adds native graph capture, and Inductor targets Rubin ( sm_107 ) For more details about these highlighted features, you can look at the release blogpost.
• Below are the full release notes for this release.
• Backwards Incompatible Changes torch.nn torch.nn.LinearCrossEntropyOptions no longer accepts acc_policy="balanced" ; use "compact" instead ( #188283 ) The "balanced" policy was removed because "compact" provides the same weight-gradient accumulation precision with lower memory use on CUDA, already uses the equivalent scratch layout for mixed-precision inputs on other devices, and was never selected by "auto" .