PyTorch Conference 2025

DeepSpeed: Efficient Training Scalability for Deep Learning

Tunji Ruwase, Snowflake

Deploying GenAI for Audio Generation on Mobile CPUs With ExecuTorch

Gian Marco Iodice, Arm

Designing and Building Custom Reinforcement Learning Environments for Fine-tuning LLMs

N. Bantilan

Distributed Training in PyTorch: Zero to Hero

Corey Lowman, Lambda Labs

Dynamic Shape Recompilations

Bob Ren, Meta

Efficient Inference Serving with Kubernetes Gateway...

Lin Sun

Efficient MoE Pre-training at Scale on AMD GPUs With TorchTitan -Liz Li & Yanyuan Qin, Matthias Reso

Empowering AI Everywhere: Democratizing PyTorch with Intel...

F. Zhao & E. Wang

Enabling Lightweight, High-Performance FSDP With NVIDIA GPU

J. Chang CN, C. Ye, X. Chen & S. Lym

Enabling PyTorch’s Thousand Ops for Software First Silicon Design

Andrew Ling

Enabling VLLM V1 on AMD GPUs With Triton

Thomas Parnell, IBM Research & Aleksandr Malyshev, AMD

Everything Everywhere all at Once: vLLM...- Brittany Rockwell & Shireen Kheradpey

ExecuTorch 1.0: General Availability Status for Mobile and Embedded...- Mergen Nachin & Cemal Bilgin

Fault Tolerance Training at Pinterest

Chia-Wei Chen & Karthik Anantha Padmanabhan, Pinterest

Fine Tune Your AI Toolbox with GitHub Copilot, Azure AI Foundry...

Sarah Kaiser

Flex Attention for Inference

Boyuan Feng & Driss Guessous, Meta

FlexCP: FlexAttention for Long Sequence Training...

Xilun Wu & Chien-Chin Huang

Foundations for AI+Science

Animashree (Anima) Anandkumar

From Model To Trust: Building Upon Tamper-proof ML Metadata Records

Mihai Maruseac, Google

FTorch: A Library To Couple PyTorch ML Models With Fortran Climate Models

M Archer

Generating State-of-the-Art GEMMs for Heterogeneous Hardware with...

Michael Lazos & Henry Tsang

Hardware-Aware Python Packages ~ PyTorch and...

Jonathan Dekhtiar & Eli Uriegas

Helion: A High-level DSL for Kernel Authoring

Jason Ansel, Meta

How Modern PyTorch Supercharges Multimodal Training and Inference at Luma AI

Thomas Neff, Luma AI

Improved GEMM and SDPA Performance on ROCm With Composable Kernel

Andres Lugo, AMD

Improving Drug Discovery with Machine Learning...

Simon Axelrod

In-Cluster Distributed Checkpointing: Optimizing Training...

G. Kroiz & S. Mishra

Jeremy Howard, Founding Researcher at fast.ai in Conversation with Anna Tong, Forbes