PyTorch Conference 2025

KernelBot: The World's First Competitive GPU Programming Platform

Mark Saroufim

Keynote Panel Discussion: Agentic AI

A. Dhinakaran, J. Liu, A. Jones, S. Colvin & B. Lorica

Keynote Panel Discussion: Signal vs. Noise: Betting on AI’s Next Wave

Keynote Panel: Hardwares & Accelerators

D Patel, S Zhou, P Salanki, N Perumbeti, M Saroufim

Kimi K2 and Our Contributions to Open Source

Yuxin Wu, Moonshot AI

Leveraging PyTorch for Generative AI in Distributed Edge Clouds

Tina Tsou

Lifecycle of a Parameter

Philip Bontrager, Meta

Mapping the Open Model Landscape in 2025

Nathan Lambert, Ai2

Marin: An Open Lab for Frontier AI

Percy Liang, Together

Maximizing Luck in Reinforcement Learning

Daniel Han, Unsloth

Model Checkpoint Compression With OpenZL

Nick Terrell & Teja Rao, Meta

Mojo + PyTorch: A Simpler, Faster Path To Custom Kernels

Spenser Bauman, Modular

Monarch: A Distributed Execution Engine for PyTorch

Colin Taylor & Zachary DeVito, Meta

Multi-Accelerator PyTorch Serving With NxD Inference and vLLM

Yahav Biran & Liangfu Chen, Amazon

mxfp8, mxfp4, nvfp4 formats and applications in PyTorch

Vasily Kuznetsov & Driss Guessous, Meta

No GPU Left Behind: Scaling Online LLM Training With...

Mert Toslali & Yu Chin Fabian Lim

Olmo-Thinking: Training a Fully Open Reasoning Model

Nathan Lambert

Open Source Model Performance Optimization With SGLang

Yineng Zhang, Together AI

Optimizing Long-Tail and MoE Challenges in Reinforcement Learning with SGLang

Chenyang Zhao, UCLA

Optimizing Model Inference with PyTorch 2.0

Devansh Ghatak

Our Journey With TorchTitan

Linsong Chu & Garrett Goon, IBM Research

Per-iteration PyTorch Visibility...

Shruti Verma & Yi-Ting Wu

Powering PyTorch: Decentralized Training for an Energy-Hungry Future

Greg Osuri

PyTorch and Democratization of AI Accelerators

Hong-Seok Kim

PyTorch APIs for High Performance MoE Training and Inference

D. Vega-Myhre; Ke Wen & N. Gimelshein

PyTorch Distributed and Fault Tolerance

Tristan Rice, Meta

PyTorch in Production: Boosting LLM Training and...

F. Hua

PyTorch Symmetric Memory: A New Programming Paradigm for Distributed AI

Ke Wen & Chien-Chin Huang