PyTorch 2.15 ships with 3,117 commits and NCCL rewrite
The NCCL rewrite becoming the default signals a maturing distributed-training stack, and NVFP4 decode targets Blackwell-class GPUs that Singapore and ASEAN data centres are actively procuring. Teams running FSDP2 at scale should review the new memory and communication controls before their next training run. Register for the 5 November Q&A to put migration questions directly to the release engineers.
PyTorch 2.15 lands with 3,117 commits from 583 contributors. The release makes the rewritten NCCL backend the default and adds a NIXL backend for one-sided transports, while FSDP2 and pipelining gain finer memory and communication control. It brings faster NVFP4 4-bit LLM decoding on NVIDIA Blackwell GPUs plus FP8 support and faster decoding on Apple Silicon. A live Q&A with Meta engineers is scheduled for 5 November at 11 AM PST.
PyTorch 2.15 is here, with 3,117 commits from 583 contributors and updates across distributed training, NVGEMM, Apple Silicon, and accelerator platform support.
The release makes the rewritten NCCL backend the default and adds one-sided transports with a NIXL backend. FSDP2 and pipelining gain finer memory and communication control. It also brings faster 4-bit (NVFP4) LLM decoding on NVIDIA Blackwell GPUs, FP8 and faster decode on Apple Silicon, and broader platform support.
On Thursday, November 5, at 11 AM PST, join Tristan Rice (@rice_fry), Natalia Gimelshein, Huy Do, and Jane Xu (@janeyx99) (@Meta) for a live PyTorch 2.15 Q&A, moderated by Chris Gottbrath (Gottbrath Tech). The speakers will share an overview of the release and answer community questions about PyTorch and the new capabilities in this release.
🔗 Read the release blog and register for the live Q&A: https://t.co/PAsbeCXk0C
