A
PyTorch Blog
@pytorch-blog
Publisher
11
Posts
Posts by PyTorch Blog
W
Towards Free Normalization: Fusing Normalization into GEMM and Attention Kernels
Code available at: https://github.com/facebookresearch/ads_model_kernel_library/tree/main/multi_cta_norm_fusion and https://github.com/facebookresearch/ads_model_kernel_library/tree/main/gdpa_megakernel TL;DR In this blog post, we present various novel kernel fusion techniques for common normaliz...
P
P
P
Understanding PyTorch’s Test Infrastructure
PyTorch tests are often generated dynamically across devices and dtypes, which is why test names in CI may look different from the class and method names in the source file. This post explains how device-generic tests, OpInfos, instantiate_device_type_tests(), and CI sharding fit together, and ho...
GD
Building the Future of On-Device AI at the ExecuTorch Hackathon
This past weekend in San Francisco, builders, researchers, mobile developers, and AI practitioners came together for the ExecuTorch Hackathon, a two-day, on-site event focused on a practical and increasingly important question: what can we build when powerful AI runs locally on the device in your...
W
Shopify Joins the PyTorch Foundation as a Platinum Member
The PyTorch Foundation, a community-driven hub for open source AI under the Linux Foundation, is announcing today that Shopify has joined as a Platinum member. Shopify provides essential internet infrastructure...
P
Miles: A PyTorch-Native Stack for Large-Scale LLM RL Post-Training
Miles is RadixArk’s open source framework for large-scale LLM RL post-training. It composes SGLang for rollout, NVIDIA Megatron-LM for training, Ray orchestration, and PyTorch-native extensibility behind a small, pluggable trainer, with unified low-precision recipes, MoE-aware rollout/training al...
D
B
TokenSpeed-Kernel: Portable APIs and High-Performance Kernels for Multi-Silicon LLM Inference
TL;DR The TokenSpeed-kernel is a standalone, open-source subsystem designed to solve backend complexity in LLM inference. It introduces a clean, layered API and registry system that decouples the high-level runtime...