Latest AI and tech news
We are excited to announce the release of PyTorch® 2.15 (release notes)!...
When you’re speaking to a device, you expect text to appear as you talk. That makes streaming automatic speech recognition (ASR) a different challenge from batch transcription, where the model has the complete recording before it starts. Strong batch...
Open source AI thrives when people come together to learn, build, share knowledge, and contribute. Behind every project is a community of developers, researchers, educators, and contributors helping make these technologies more accessible to others....
PyTorch Conference North America 2026 brought 3,500+ attendees to San Jose on October 20–21 for…
TorchTitan is a PyTorch-native platform for rapid experimentation and large-scale pre- and post-training of generative…
PyTorch users on AMD MI350-series GPUs can now use FlyDSL through existing operator APIs for…
New Platinum members include Nebius, Nscale and Shopify, deepening the PyTorch Foundation’s role as the intelligence layer for open source AI development...
The PyTorch Foundation, the vendor-neutral home for the open source intelligence layer, is announcing today that Nebius has joined as a Platinum member....
TorchTPU is an open source collaboration between Google and Meta that introduces a native and principled PyTorch backend for Google TPUs. It enables seamless out-of-the-box model execution and TPU migration while delivering industry-competitive infer...
Torch-spyre’s Inductor decides how each kernel’s work is divided across Spyre’s cores and manages tensor residency between on-chip scratchpad and off-chip memory, and the backend compiler turns those decisions into something the hardware can actually...
London, U. K., 20 October 2026 – The PyTorch Foundation, the vendor-neutral home for the open source intelligence layer, is announcing today that Nscale has joined as a Platinum member. Nscale is a full-stack AI cloud platform, bringing together powe...
Frontier models span language, video and audio, on architectures from dense and mixture-of-experts transformers to diffusion, state-space, world and structure-prediction models. PyTorch is where AI research happens, where the models are defined, and ...
With Helion’s release last year, it aimed at enabling developers to write GPU kernels in…
Join us in Tokyo on December 10 for a full day dedicated to the open…
The data that a model sees during training is one of the most powerful levers we have for influencing how well it performs. Much of the day-to-day work of improving a model happens on the data side: we change the mixture, or the filters that we apply...
Today, Meta is releasing the weights for Muse Spark Open, the most capable US open-weights model to date. It’s available today on Hugging Face under the Muse Model License. Muse Spark Open runs optimized on the SGLang and vLLM inference engines, with...
Microcontrollers power billions of always-on devices: wearables, industrial sensors, voice interfaces, and smart home appliances. These devices operate under extreme constraints: often less than 1 MB of SRAM, single-digit megabytes of flash, and powe...
The PyTorch-based software stack for IBM’s Spyre accelerator recently moved its torch.compile backend from sendnn, a separate compilation stack built entirely from scratch, to backend=inductor, an open-source backend built by extending PyTorch’s own ...
Agentic workloads change the traffic an inference server sees. Unlike single-turn chat, an agent session can involve a large initial prefill, repeated model calls, and subagents running in parallel, with KV cache remaining resident while tools execut...
Spyre becomes a native PyTorch device by connecting PyTorch’s existing device, allocator, stream, and compiler abstractions through torch-spyre to the Spyre runtime and firmware. PrivateUse1 gives Spyre a real device identity, PyTorch allocator stora...
This post explores the FBTriton kernel design for Table Batched Embedding (TBE) forward and backward passes. These core operators handle embedding lookups across thousands of sharded GPUs within recommendation systems. Our Triton implementation succe...
The Accelerator Integration Working Group plays a vital role in standardizing how new hardware architectures connect to the open source AI ecosystem. As compute platforms diversify across cloud, edge, and specialized silicon, the Accelerator Integrat...
TL;DR If you need to decode or encode media, whether it's images, video, or audio,…
We integrated Helion into vLLM’s linear backend to explore how an autotuned, high-level kernel DSL can improve LLM inference performance while reducing kernel implementation complexity. A single Helion general matrix multiplication (GEMM) implementat...
We’re excited to bring you the new PyTorch Certified Associate (PTCA) Certification Pathway, combining focused learning modules and the PyTorch Certified Associate (PTCA) certification exam together in one structured experience through Linux Foundati...
TL;DR In this blog post, we present our work on Jagged Flash Attention (JFA) —…
TL;DR In only a few weeks, PyTorch Conference North America 2026 will begin in San…
TL;DR PyTorch's Cross-Repository CI Relay (CRCR) gives out-of-tree accelerators a clean, scalable way to plug…
As deep learning models move rapidly from research prototypes into core enterprise infrastructure, the demand for practical, end-to-end PyTorch expertise has never been higher. Building robust neural networks requires more than just high-level unders...
Some of the most interesting work being built with PyTorch starts in universities, research labs,…
To achieve state-of-the-art performance at the frontier, vLLM is changing its internal implementation in ways that make it incompatible with fullgraph torch.compile. This may have consequences for users who care about out-of-tree accelerators, older ...
TL;DR: This case study explores how Shopify compresses production failures into model weights every day, beats frontier-model quality, and cuts serving costs 96% by building a continual learning loop with PyTorch and vLLM....
A framework you write yourself, tensors through transformers TL;DR Every mature systems project eventually needs…
PyTorch Day Japan 2026 will bring the open source AI community together in Tokyo on December 10 for a full day of technical talks and interactive discussions designed to foster knowledge exchange and collaboration....
Taking place October 20 to 21 in San Jose, California, PyTorch Conference North America 2026 highlights open research, tooling, and performance optimization across compiler architecture, cross-hardware kernel domain-specific languages, exascale distr...
We extend FlashAttention-4 [1] with MXFP8 forward and backward, reaching 2.85 PF/s forward and 2 PF/s backward on LLM shapes. On our internal shapes, FA4 MX8 reaches 2.54 PF/s forward and 1.58 PF/s backward, delivering up to 1.6× and 1.52× gains over...
TL;DR The HuggingFace Kernels project now has Helion support. This blog walks through how to…
PyTorch Conference China 2026 brought the PyTorch community together in Shanghai on September 8–9 alongside…
China’s leading AI technology companies Alibaba Cloud, Cambricon and Ant Group join the PyTorch Foundation as members...
The PyTorch Foundation, a community-driven hub for open source AI under the Linux Foundation, is announcing today that Cambricon has joined as a Platinum member....
More than 170 students, engineers, researchers, and open-source contributors gathered in Bengaluru for a technical evening hosted by Red Hat and Hugging Face around PyTorch, large-scale inference, reinforcement learning environments, distributed trai...
PyTorch Conference North America 2026 (San Jose, October 20–21) is packed with sessions on getting PyTorch to run fast, portably, and reliably across an increasingly diverse silicon landscape – GPUs, TPUs, NPUs, and custom ASICs alike....