@PyTorch: PyTorch 2.14 introduces updates to Inductor, PyTorch Distributed, c10d, dynamic shapes, Apple Silicon, and accelerator …

X AI KOLs Timeline Tools
pytorch deep-learning framework-update inductor distributed-computing apple-silicon compiler

Summary

PyTorch 2.14 introduces updates to Inductor, PyTorch Distributed, dynamic shapes, Apple Silicon support, and more, with a live Q&A and PyTorchCon announcement.

PyTorch 2.14 introduces updates to Inductor, PyTorch Distributed, c10d, dynamic shapes, Apple Silicon, and accelerator platform support, with 2,995 commits from 487 contributors since PyTorch 2.13. Bring your questions about PyTorch 2.14 to our live Q&A on Thursday, September 17, 2026. Andrey Talman (@Meta), Natalia Gimelshein (@Meta), @joespeez (@reflection_ai), and Chris Gottbrath (Gottbrath Tech, moderator) will share an overview of PyTorch 2.14 and answer community questions about PyTorch and the new capabilities. Register: https://streamyard.com/watch/E7iTfnUvpvTW… PyTorch 2.14 covers: R: UTLASS kernels in Inductor - The new nccl2 backend for PyTorch Distributed - Fault-tolerant collectives and process-group reconfiguration in c10d - Native linear algebra and additional Metal kernel improvements on Apple Silicon - torch.switch and CUDA graph capture for torch.while_loop - Declarative dynamic shapes with @dynamic_spec - Experimental torch.compile support for complex-valued tensors - Expanded ROCm, Intel XPU, and NVIDIA platform support Want to learn more about compiler and runtime work, distributed communication, device portability, accelerator integration, and more? Register for #PyTorchCon North America 2026, where engineers, researchers, and maintainers will convene October 20–21 in San Jose: https://hubs.la/Q04tDbKG0
Original Article
View Cached Full Text

Cached at: 09/02/26, 06:24 AM

PyTorch 2.14 introduces updates to Inductor, PyTorch Distributed, c10d, dynamic shapes, Apple Silicon, and accelerator platform support, with 2,995 commits from 487 contributors since PyTorch 2.13.

Bring your questions about PyTorch 2.14 to our live Q&A on Thursday, September 17, 2026. Andrey Talman (@Meta), Natalia Gimelshein (@Meta), @joespeez (@reflection_ai), and Chris Gottbrath (Gottbrath Tech, moderator) will share an overview of PyTorch 2.14 and answer community questions about PyTorch and the new capabilities.

Register: https://streamyard.com/watch/E7iTfnUvpvTW…

PyTorch 2.14 covers: R: UTLASS kernels in Inductor

  • The new nccl2 backend for PyTorch Distributed
  • Fault-tolerant collectives and process-group reconfiguration in c10d
  • Native linear algebra and additional Metal kernel improvements on Apple Silicon
  • torch.switch and CUDA graph capture for torch.while_loop
  • Declarative dynamic shapes with @dynamic_spec
  • Experimental torch.compile support for complex-valued tensors
  • Expanded ROCm, Intel XPU, and NVIDIA platform support

Want to learn more about compiler and runtime work, distributed communication, device portability, accelerator integration, and more? Register for #PyTorchCon North America 2026, where engineers, researchers, and maintainers will convene October 20–21 in San Jose: https://hubs.la/Q04tDbKG0

Similar Articles