Blog – PyTorch

May 20, 2025

Accelerating GPU Performance with Triton: April 30th PyTorch ATX Event

May 13, 2025

PyTorch/XLA 2.7 Release Usability, vLLM boosts, JAX bridge, GPU Build

May 12, 2025

MetaShuffling: Accelerating Llama 4 MoE Inference

May 7, 2025

PyTorch: The Open Language of AI

May 5, 2025

Recap of the PyTorch Korea User Group Meetup: A Technical Conference with a PyTorch Core Maintainer

Apr 30, 2025

FlexAttention Part II: FlexAttention for Inference

Apr 30, 2025

6x faster Async Checkpointing in PyTorch, using Cached Plans, no GIL contention

Apr 28, 2025

Accelerating Large Scale Training and Convergence with PyTorch Float8 Rowwise on Crusoe 2K H200s

Apr 25, 2025

Accelerate PyTorch 2.7 on Intel® GPUs

Apr 23, 2025

PyTorch 2.7 Release

Apr 8, 2025

Accelerating Whisper on Arm with PyTorch and Hugging Face Transformers

Mar 19, 2025

PyTorch Day China 2025 Call for Proposals Open

Mar 11, 2025

Scaling Recommendation Systems Training to Thousands of GPUs with 2D Sparse Parallelism

Mar 6, 2025

Peak Performance, Minimized Memory: Optimizing torchtune’s performance with torch.compile & Liger Kernel

Mar 5, 2025

Current and New Activation Checkpointing Techniques in PyTorch

Feb 26, 2025

Accelerating Generative AI with PyTorch: Segment Anything 2 – Fast and furious inference with low latency and fast cold starts

Feb 11, 2025

Unlocking the Latest Features in PyTorch 2.6 for Intel Platforms

Feb 5, 2025

Enabling advanced GPU features in PyTorch – Warp Specialization

Jan 29, 2025

PyTorch 2.6 Release Blog

Jan 28, 2025

2025 Priorities for the PyTorch Technical Advisory Council (TAC)

Jan 24, 2025

How Intel Uses PyTorch to Empower Generative AI through Intel Arc GPUs

Jan 21, 2025

Accelerating LLM Inference with GemLite, TorchAO and SGLang

Jan 14, 2025

GenAI Acceleration for PyTorch 2.5 on Intel® Xeon®Processors

Jan 9, 2025

Integrating Ascend Backend with Torchtune through PyTorch Multi-Device Support

Jan 6, 2025

High-Performance Low-Bit Operators for PyTorch

Dec 20, 2024

Improve RAG performance with torch.compile on AWS Graviton Processors

Dec 11, 2024

torchcodec: Easy and Efficient Video Decoding for PyTorch

Dec 6, 2024

Accelerating 2D Dynamic Block Quantized Float8 GEMMs in Triton

Dec 2, 2024

HadaCore: Tensor Core Accelerated Hadamard Transform Kernel

Nov 25, 2024

Supercharging Training using float8 and FSDP2

Nov 18, 2024

Distilling Llama3.1 8B into 1B in torchtune

Nov 1, 2024

Deep Dive on CUTLASS Ping-Pong GEMM Kernel

Oct 31, 2024

Deploying LLMs with TorchServe + vLLM

Oct 30, 2024

Triton Kernel Compilation Stages

Oct 28, 2024

Unleashing the Power of AI on Mobile: LLM Inference for Llama 3.2 Quantized Models with ExecuTorch and KleidiAI

Oct 28, 2024

Getting started with PyTorch, ExecuTorch, and Ethos-U85 in three easy steps

Oct 25, 2024

Intel GPU Support Now Available in PyTorch 2.5

Oct 24, 2024

ExecuTorch Beta: On-Device AI and LLMs, Stability, and Acceleration with Partners

Oct 23, 2024

TorchRec and FBGEMM 1.0 Stable Release

Oct 17, 2024

PyTorch 2.5 Release Blog

Oct 15, 2024

The Path to Achieve PyTorch Performance Boost on Windows CPU

Oct 8, 2024

PyTorch Foundation Technical Advisory Council Elects New Leadership

Sep 18, 2024

Challenges and Efforts in PyTorch Multi-Device Integration: Compatibility, Portability, and Integration Efficiencies

Sep 4, 2024

CUDA-Free Inference for LLMs

Aug 29, 2024

Accelerate Your AI: PyTorch 2.4 Now Supports Intel GPUs for Faster Workloads

Aug 20, 2024

Enabling Fast Gradient Clipping and Ghost Clipping in Opacus

Aug 7, 2024

FlexAttention: The Flexibility of PyTorch with the Performance of FlashAttention

Jul 30, 2024

Quantization-Aware Training for Large Language Models with PyTorch

Jul 24, 2024

PyTorch 2.4 Release Blog

Jul 22, 2024

Deep Dive on the Hopper TMA Unit for FP8 GEMMs

Jul 11, 2024

FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision

Jul 10, 2024

Learn how to develop Android applications with ExecuTorch and Llama models

Jul 9, 2024

Accelerated PyTorch inference with torch.compile on AWS Graviton processors

Jul 3, 2024

Announcing Hacker Cup AI Track at NeurIPS 2024

Jun 23, 2024

Training MoEs at Scale with PyTorch

Jun 20, 2024

Accelerating Neural Network Training with Semi-Structured (2:4) Sparsity

Jun 12, 2024

Reducing Model Checkpointing Times by Over 10x with PyTorch Distributed Asynchronous Checkpointing

Jun 6, 2024

INT4 Decoding GQA CUDA Optimizations for LLM Inference

May 21, 2024

Maximizing Training Throughput Using PyTorch FSDP and Torch.compile

May 15, 2024

Achieving Sustainability Goals with PyTorch and Intel AI

May 14, 2024

Speeding up ViTs using Block Sparsity

May 2, 2024

A Hitchhiker’s Guide to Speculative Decoding

May 1, 2024

Accelerating Llama3 FP8 Inference with Triton Kernels

Apr 30, 2024

ExecuTorch Alpha: Taking LLMs and AI to the Edge with Our Community and Partners

Apr 24, 2024

PyTorch 2.3 Release Blog

Apr 4, 2024

Accelerating MoE model inference with Locality-Aware Kernel Design

Mar 13, 2024

Maximizing training throughput using PyTorch FSDP

Jan 23, 2024

Accelerating Generative AI with PyTorch IV: Seamless M4T, fast

Jan 18, 2024

Accelerate PyTorch Models Using Quantization Techniques with Intel Extension for PyTorch

Jan 16, 2024

Accelerating Triton Dequantization Kernels for GPTQ

Jan 10, 2024

Finetune LLMs on your own consumer hardware using tools from PyTorch and Hugging Face ecosystem

Jan 9, 2024

Accelerate AI models on GPU using Amazon SageMaker multi-model endpoints with TorchServe, saving up to 75% on inference costs

Jan 3, 2024

Accelerating Generative AI Part III: Diffusion, Fast

Dec 19, 2023

Understanding GPU Memory 2: Finding and Removing Reference Cycles

Dec 18, 2023

Training Production AI Models with PyTorch 2.0

Dec 15, 2023

Empowering Models with Performance: The Art of Generalized Model Transformation Approach

Dec 14, 2023

Understanding GPU Memory 1: Visualizing All Allocations over Time

Dec 12, 2023

From PyTorch Conference 2023: From Dinosaurs to Seismic Imaging with Intel

Nov 30, 2023

Accelerating Generative AI with PyTorch II: GPT, Fast

Nov 29, 2023

PyTorch 2.1 Contains New Performance Features for AI Developers

Nov 16, 2023

Accelerating Generative AI with PyTorch: Segment Anything, Fast

Nov 7, 2023

PyTorch compile to speed up inference on Llama 2

Nov 6, 2023

High-Performance Llama 2 Training and Inference with PyTorch/XLA on Cloud TPUs

Nov 2, 2023

Accelerating Inference on x86-64 Machines with oneDNN Graph

Oct 31, 2023

AMD Extends Support for PyTorch Machine Learning Development on Select RDNA™ 3 GPUs with ROCm™ 5.7

Oct 17, 2023

Compiling NumPy code into C++ or CUDA via torch.compile

Oct 13, 2023

Flash-Decoding for long-context inference

Oct 11, 2023

ML Model Server Resource Saving – Transition From High-Cost GPUs to Intel CPUs and oneAPI powered Software with performance

Oct 10, 2023

Real-time Audio-visual Speech Recognition

Oct 4, 2023

PyTorch 2.1: automatic dynamic shape compilation, distributed checkpointing

Oct 4, 2023

New Library Updates in PyTorch 2.1

Oct 4, 2023

High performance Llama 2 deployments with AWS Inferentia2 using TorchServe

Oct 3, 2023

How to Build an Interactive Chat-Generation Model using DialoGPT and PyTorch

Sep 25, 2023

Inside the Matrix: Visualizing Matrix Multiplication, Attention and Beyond

Sep 13, 2023

Accelerated CPU Inference with PyTorch Inductor using torch.compile

Sep 5, 2023

Automated trace collection and analysis

Aug 31, 2023

PyTorch/XLA SPMD: Scale Up Model Training and Serving with Automatic Parallelization

Aug 24, 2023

Large Scale Training of Hugging Face Transformers on TPUs With PyTorch/XLA FSDP

Aug 7, 2023

INT8 Quantization for x86 CPU in PyTorch

Jul 25, 2023

Announcing CPP-based S3 IO DataPipes

Jul 10, 2023

How to Accelerate PyTorch Geometric on Intel® CPUs

Jun 29, 2023

Optimizing LibTorch-based inference engine memory usage and thread-pooling

Jun 28, 2023

The Path to Achieve Ultra-Low Inference Latency With LLaMA 65B on PyTorch/XLA

Jun 22, 2023

Optimized PyTorch 2.0 Inference with AWS Graviton processors

Jun 16, 2023

🎉 PyTorch Docathon H1 2023 Wrap-up 🎉

May 22, 2023

Out of the box acceleration and memory savings of 🤗 decoder models with PyTorch 2.0

May 12, 2023

Language Identification: Building an End-to-End AI Solution using PyTorch

May 2, 2023

Accelerated Image Segmentation using PyTorch

Apr 27, 2023

Introducing Hidet: A Deep Learning Compiler for Efficient Model Serving

Apr 19, 2023

Accelerating Large Language Models with Accelerated Transformers

Apr 15, 2023

Experience the power of PyTorch 2.0 on AMD Solutions

Apr 14, 2023

Accelerated Generative Diffusion Models with PyTorch 2

Apr 7, 2023

Straggler Mitigation On PyTorch DDP By Hierarchical SGD

Apr 7, 2023

Celebrate PyTorch 2.0 with New Performance Features for AI Developers

Apr 3, 2023

PyTorch & OpenXLA: The Path Forward

Mar 28, 2023

Accelerated PyTorch 2 Transformers

Mar 22, 2023

PyTorch 2.0 & XLA—The Latest Cutting Edge Features

Mar 16, 2023

Accelerated Diffusers with PyTorch 2.0

Mar 15, 2023

PyTorch 2.0: Our next generation release that is faster, more Pythonic and Dynamic as ever

Mar 15, 2023

New Library Updates in PyTorch 2.0

Feb 14, 2023

Democratizing AI with PyTorch Foundation and ROCm™ support for PyTorch

Feb 2, 2023

Deprecation of CUDA 11.6 and Python 3.7 Support

Jan 9, 2023

PyTorch Trace Analysis for the Masses

Dec 31, 2022

Compromised PyTorch-nightly dependency chain between December 25th and December 30th, 2022.

Dec 28, 2022

Torchserve Performance Tuning, Animated Drawings Case-Study

Dec 22, 2022

Scaling Vision Model Training Platforms with PyTorch

Dec 16, 2022

Efficient Large-Scale Training with Pytorch FSDP and AWS

Dec 15, 2022

Scaling PyTorch FSDP for Training Foundation Models on IBM Cloud

Dec 2, 2022

Get Started with PyTorch 2.0 Summary and Overview

Dec 2, 2022

Accelerating Hugging Face and TIMM models with PyTorch 2.0

Nov 28, 2022

Optimizing Production PyTorch Models’ Performance with Graph Transformations

Nov 22, 2022

Efficient Multi-Objective Neural Architecture Search with Ax

Nov 21, 2022

Scaling Multimodal Foundation Models in TorchMultimodal with Pytorch Distributed

Nov 17, 2022

Introducing TorchMultimodal – a library for accelerating exploration in Multimodal AI

Nov 10, 2022

PyTorch Enterprise Support Program Update

Nov 3, 2022

Extending TorchVision’s Transforms to Object Detection, Segmentation & Video tasks

Oct 28, 2022

New Library Updates in PyTorch 1.13

Oct 28, 2022

PyTorch 1.13 release, including beta versions of functorch and improved support for Apple’s new M1 chips.

Oct 17, 2022

PyTorch’s Tracing Based Selective Build

Oct 13, 2022

Scaling PyTorch models on Cloud TPUs with FSDP

Sep 29, 2022

Performance Debugging of Production PyTorch Models at Meta

Sep 26, 2022

Announcing PyTorch Conference 2022

Sep 12, 2022

PyTorch strengthens its governance by joining the Linux Foundation

Aug 29, 2022

Fast Beam Search Decoding in PyTorch with TorchAudio and Flashlight Text

Aug 26, 2022

Introducing nvFuser, a deep learning compiler for PyTorch

Aug 24, 2022

Accelerating PyTorch Vision Models with Channels Last on CPU

Aug 18, 2022

Easily list and initialize models with new APIs in TorchVision

Aug 16, 2022

Empowering PyTorch on Intel® Xeon® Scalable processors with Bfloat16

Jul 22, 2022

Introducing the PlayTorch app: Rapidly Create Mobile AI Experiences

Jul 19, 2022

What Every User Should Know About Mixed Precision Training in PyTorch

Jul 15, 2022

Case Study: PathAI Uses PyTorch to Improve Patient Outcomes with AI-powered Pathology

Jul 12, 2022

A BetterTransformer for Fast Transformer Inference

Jun 28, 2022

PyTorch 1.12: TorchArrow, Functional API for Modules and nvFuser, are now available

Jun 28, 2022

New library updates in PyTorch 1.12

Jun 27, 2022

How Computational Graphs are Executed in PyTorch

Jun 23, 2022

Geospatial deep learning with TorchGeo

Jun 16, 2022

How Disney Improved Activity Recognition Through Multimodal Approaches with PyTorch

May 18, 2022

Introducing Accelerated PyTorch Training on Mac

May 12, 2022

Ambient Clinical Intelligence: Generating Medical Reports with PyTorch

Mar 16, 2022

Running PyTorch Models on Jetson Nano

Mar 14, 2022

Introducing PyTorch Fully Sharded Data Parallel (FSDP) API

Mar 10, 2022

PyTorch 1.11, TorchData, and functorch are now available

Mar 10, 2022

Introducing TorchRec, and other domain library updates in PyTorch 1.11

Mar 2, 2022

Understanding LazyTensor System Performance with PyTorch/XLA on Cloud TPU

Feb 24, 2022

Case Study: Amazon Ads Uses PyTorch and AWS Inferentia to Scale Models for Ads Processing

Feb 23, 2022

Introducing TorchRec, a library for modern production recommendation systems

Feb 8, 2022

Practical Quantization in PyTorch

Dec 22, 2021

Introducing TorchVision’s New Multi-Weight Support API

Dec 15, 2021

Efficient PyTorch: Tensor Memory Format Matters

Dec 8, 2021

Announcing the Winners of the 2021 PyTorch Annual Hackathon

Nov 18, 2021

How to Train State-Of-The-Art Models Using TorchVision’s Latest Primitives

Oct 29, 2021

Feature Extraction in TorchVision using Torch FX

Oct 26, 2021

Accelerating PyTorch with CUDA Graphs

Oct 21, 2021

PyTorch 1.10 Release, including CUDA Graphs APIs, Frontend and Compiler Improvements

Oct 21, 2021

New Library Releases in PyTorch 1.10, including TorchX, TorchAudio, TorchVision

Sep 8, 2021

Announcing PyTorch Annual Hackathon 2021

Aug 31, 2021

How Computational Graphs are Constructed in PyTorch

Aug 23, 2021

Announcing PyTorch Developer Day 2021

Aug 18, 2021

PipeTransformer: Automated Elastic Pipelining for Distributed Training of Large-scale Models

Aug 3, 2021

What’s New in PyTorch Profiler 1.9?

Jun 27, 2021

Everything You Need To Know About Torchvision’s SSDlite Implementation

Jun 23, 2021

The torch.linalg module: Accelerated Linear Algebra with Autograd in PyTorch

Jun 18, 2021

An Overview of the PyTorch Mobile Demo Apps

Jun 16, 2021

Everything You Need To Know About Torchvision’s SSD Implementation

Jun 15, 2021

New PyTorch Library Releases in PyTorch 1.9, including TorchVision, TorchAudio, and more

Jun 15, 2021

PyTorch 1.9 Release, including torch.linalg and Mobile Interpreter

Jun 8, 2021

Overview of PyTorch Autograd Engine

May 26, 2021

Everything you need to know about TorchVision’s MobileNetV3 implementation

May 25, 2021

Announcing the PyTorch Enterprise Support Program

May 10, 2021

PyTorch Ecosystem Day 2021 Recap and New Contributor Resources

Apr 16, 2021

An overview of the ML models introduced in TorchVision v0.9

Mar 25, 2021

Introducing PyTorch Profiler – the new and improved performance tool

Mar 24, 2021

PyTorch for AMD ROCm™ Platform now available as Python package

Mar 9, 2021

Announcing PyTorch Ecosystem Day

Mar 4, 2021

New PyTorch library releases including TorchVision Mobile, TorchAudio I/O, and more

Mar 4, 2021

PyTorch 1.8 Release, including Compiler and Distributed Training updates, and New Mobile Tutorials

Mar 3, 2021

The torch.fft module: Accelerated Fast Fourier Transforms with Autograd in PyTorch

Nov 12, 2020

Prototype Features Now Available – APIs for Hardware Accelerated Mobile and ARM64 Builds

Nov 1, 2020

Announcing PyTorch Developer Day 2020

Oct 27, 2020

PyTorch 1.7 released w/ CUDA 11, New APIs for FFTs, Windows support for Distributed training and more

Oct 1, 2020

Announcing the Winners of the 2020 Global PyTorch Summer Hackathon

Aug 24, 2020

PyTorch framework for cryptographically secure random number generation, torchcsprng, now available

Aug 18, 2020

PyTorch 1.6 now includes Stochastic Weight Averaging

Aug 11, 2020

Efficient PyTorch I/O library for Large Datasets, Many Files, Many GPUs

Jul 28, 2020

PyTorch 1.6 released w/ Native AMP Support, Microsoft joins as maintainers for Windows

Jul 28, 2020

PyTorch feature classification changes

Jul 28, 2020

Microsoft becomes maintainer of the Windows version of PyTorch

Jul 28, 2020

Introducing native PyTorch automatic mixed precision for faster training on NVIDIA GPUs

May 5, 2020

Updates & Improvements to PyTorch Tutorials

Apr 21, 2020

PyTorch 1.5 released, new and updated APIs including C++ frontend API parity with Python

Apr 21, 2020

PyTorch library updates including new model serving library

Mar 26, 2020

Introduction to Quantization on PyTorch

Jan 15, 2020

PyTorch 1.4 released, domain libraries updated

Dec 6, 2019

OpenMined and PyTorch partner to launch fellowship funding for privacy-preserving ML community

Nov 16, 2019

PyTorch adds new tools and libraries, welcomes Preferred Networks to its community

Oct 10, 2019

PyTorch 1.3 adds mobile, privacy, quantization, and named tensors

Aug 8, 2019

New Releases: PyTorch 1.2, torchtext 0.4, torchaudio 0.3, and torchvision 0.4

Jul 23, 2019

Mapillary Research: Seamless Scene Segmentation and In-Place Activated BatchNorm

Jul 18, 2019

PyTorch Adds New Ecosystem Projects for Encrypted AI and Quantum Computing, Expands PyTorch Hub

Jun 10, 2019

Towards Reproducible Research with PyTorch Hub

May 22, 2019

torchvision 0.3: segmentation, detection models, new datasets and more..

May 8, 2019

Model Serving in PyTorch

May 1, 2019

Optimizing CUDA Recurrent Neural Networks with TorchScript

May 1, 2019

PyTorch adds new dev tools as it hits production scale

Apr 29, 2019

Stochastic Weight Averaging in PyTorch

May 2, 2018

The road to 1.0: production ready PyTorch

Apr 22, 2018

PyTorch 0.4.0 Migration Guide

Mar 5, 2018

Tensor Comprehensions in PyTorch

Jan 19, 2018

PyTorch, a year in….

Jun 27, 2017

PyTorch Internals Part II – The Build System

May 11, 2017

A Tour of PyTorch Internals (Part I)

Docs

Access comprehensive developer documentation for PyTorch

Tutorials

Get in-depth tutorials for beginners and advanced developers

View Tutorials ›

Resources

Find development resources and get your questions answered

View Resources ›

Stay in touch for updates, event info, and the latest news

By submitting this form, I consent to receive marketing emails from the LF and its projects regarding their events, training, research, developments, and related announcements. I understand that I can unsubscribe at any time using the links in the footers of the emails I receive. Privacy Policy.

© 2025 PyTorch. Copyright © The Linux Foundation®. All rights reserved. The Linux Foundation has registered trademarks and uses trademarks. For more information, including terms of use, privacy policy, and trademark usage, please see our Policies page. Trademark Usage. Privacy Policy.