PyTorch

PyTorch Helion Adds Compile-Time Heuristics to Cut Autotuning Delays
PyTorch’s Helion now uses compile-time heuristics to select kernel settings without running slow autotuning searches. The system analyzes a kernel’s intermediate…

PyTorch team integrates SGLang with IBM Spyre in less than a week
A two-person team integrated SGLang with IBM Spyre in less than a week, using AI coding agents to build an open source plugin of roughly 1,500 lines. PyTorch says SGLang…

PyTorch and TorchTitan Train a 1T-Parameter MoE on JUPITER
The EUROPA consortium says it has scaled pretraining for a roughly 1-trillion-parameter Mixture-of-Experts model on JUPITER, a publicly funded European supercomputer.…

Meta Introduces TLX, a Hardware-Native Compiler for GPU Workloads
Meta has detailed Triton Low-Level Extensions, or TLX, an extension to Triton designed to improve GPU performance in production machine learning workloads. TLX gives…

PyTorch Models Can Now Run on GPUs Inside Web Browsers
PyTorch models can now run on a device’s GPU inside the web browser through the new ExecuTorch WebGPU backend. The approach uses a single ExecuTorch .pte file, avoids…

PyTorch serves Meta’s 816B Muse Spark Open on one TPU v7 host
PyTorch says it can serve Meta’s 816B-parameter Muse Spark Open mixture-of-experts model on a single TPU v7 host containing four chips. The system uses SGLang-TorchTPU…

Databricks beta streams Unity Catalog data directly to GPUs
Databricks has released UCDataset in beta on AI Runtime, a native PyTorch data loader that streams Unity Catalog tables directly to GPUs. The company says it can cut…

Google makes Pathways generally available for large-model training
Google has announced the general availability of Pathways on Google Cloud Platform, a distributed system it built to train large models including Gemini. Pathways uses a…

vLLM adopts LibTorch Stable ABI to ease PyTorch upgrades
vLLM has migrated its CUDA build to PyTorch’s LibTorch Stable ABI, allowing its v0.33.0 LibTorch ABI Stable CUDA wheels to work across compatible PyTorch releases…

Ray Launches Special Interest Group to Shape Accelerator Support
The Ray community has created a Special Interest Group (SIG) for hardware accelerators, aiming to define a shared, vendor-neutral roadmap for how Ray works with GPUs…

PyTorch 2.15 makes rewritten NCCL backend the default and adds FP8 on Apple Silicon
PyTorch 2.15 is out, and the rewritten NCCL backend now runs every nccl process group by default. The release also adds one-sided transports to torch.distributed, finer…
Arm SME2 Cuts Streaming Speech Recognition Time on Pixel 11 by Up to 2.12x
A PyTorch case study finds that Arm's SME2 matrix extension cuts whole-recording processing time for the streaming speech model Nemotron 3.5 on a Google Pixel 11 by…

PyTorch Conference 2026: TPU, Trainium Backends and Nine New Members
PyTorch Conference North America 2026 in San Jose drew more than 3,500 attendees and brought announcements on native PyTorch backends for Google TPUs and AWS Trainium…

PyTorch Foundation names 24 ambassadors for 2026 across five regions
The PyTorch Foundation has named its 2026 Ambassador Program cohort: 24 community leaders across five program regions. According to the Foundation, they will run…

Vultr joins Agentic AI Foundation as Gold member to push open agent stacks
Cloud infrastructure provider Vultr has joined the Agentic AI Foundation (AAIF) as a Gold member. The company says it will support the foundation's work on open…

TorchTitan adds RL training with TitanRL and support for new frontier models
PyTorch's TorchTitan platform has gained reinforcement learning support through a new component called TitanRL, along with native support for several new open-weights…

PyTorch adds FlyDSL backend for faster operators on AMD MI350 GPUs
PyTorch users on AMD MI350-series GPUs can now use FlyDSL kernels through existing operator APIs for GEMM, attention, RMSNorm and TopK. According to the PyTorch blog…

i14 brings Laplace 3B linear attention model to AWS Trainium2 with native PyTorch
AI research company i14 has trained and served its Laplace 3B linear attention model on AWS Trainium2 using native PyTorch, according to an AWS blog post. The model…

PyTorch Foundation adds nine members, including Nebius, Nscale and Shopify
The PyTorch Foundation, the Linux Foundation-hosted hub for open source AI, has added nine new members, with Nebius, Nscale and Shopify joining at the Platinum tier. The…

Nebius joins PyTorch Foundation as Platinum member, gets a board seat
Nebius, the Amsterdam-based AI cloud company, has joined the PyTorch Foundation as a Platinum member, according to a foundation announcement. The membership comes with a…

Google and Meta launch TorchTPU, a native PyTorch backend for TPUs
Google and Meta's PyTorch team have introduced TorchTPU, an open source backend that runs PyTorch natively on Google TPUs. According to the project's authors, existing…

PyTorch Team Details MLIR Dataflow Backend That Turns KTIR Into IBM Spyre Code
A new PyTorch blog post describes how torch-spyre turns Inductor's work-division decisions into programs for IBM's Spyre accelerator. The path runs through KTIR, an MLIR…

Ray now runs native PyTorch on Google TPUs, Anyscale and Google Cloud say
Anyscale and Google Cloud have announced native Ray support for PyTorch on Google TPUs. Built on Google's TorchTPU project and KubeRay on Google Kubernetes Engine (GKE)…

Hugging Face adds native TPU support through TorchTPU PyTorch backend
Hugging Face says Transformers, Diffusers and Kernels now run natively on Google's TPUs, thanks to TorchTPU, a native PyTorch backend that makes TPU "just another…