11:10am • Lightning Talk: What’s New in Export? - Angela Yi, Tugsbayasgalan Manlaibaatar, Avik Chaudhuri & Yidi Wu, Meta
11:25am • Lightning Talk: Low Precision Dtypes in PyTorch - Vasiliy Kuznetsov, Meta
11:40am • ExecuTorch Beta and on-Device Generative AI Support - Mergen Nachin & Mengtao (Martin) Yuan, Meta
2:10pm • State of PyTorch - Ji Li & Damien Sereni, Meta
2:40pm • Sponsored Session: Accelerating AI Innovation: High Performance PyTorch at AMD - Robert Suderman & Ian Norden, AMD
3:10pm • Lightning Talk: A Whirlwind Tour of PyTorch Extension Points - Alban Desmaison, Meta
3:25pm • Lightning Talk: Extending PyTorch with Custom Python/C++/CUDA Operators - Richard Zou, Meta
4:00pm • Welcome to the PyTorch Ecosystem for LLM Fine-tuning Mini Summit - Kartikay Khandelwal, Meta
4:05pm • The State of the Llama Ecosystem - Joe Spisak, Meta
4:15pm • The Challenges of Building an Opinionated Open Source LLM Framework - Wing Lian, Axolotl AI
4:25pm • Hacks to Make LLM Training Faster - Daniel Han, Unsloth AI
4:35pm • torchtune: Easy and Accessible Finetuning in Native PyTorch - Evan Smothers, Meta
4:45pm • Panel Discussion - Tim Dettmers, AI2/Carnegie Melon; Hailey Schoelkopf, EleutherAI; Aakanksha Chowdhery, Meta; Alexis Conneau, OpenAI; Moderated by Kartikay Khandelwal, Meta
11:10am • Sponsored Session: NeMo-Aligner: A Scalable Toolkit for Model Alignment - Gerald Shen & Jimmy Zhang, NVIDIA
11:40am • Lightning Talk: HieroGlyph2Text: A PyTorch-Powered Pipeline for Automated Egyptian Hieroglyph Translation from Image - Susi Gentsch, University of Bonn
11:55am • Lightning Talk: Mobile Computational Photography with PyTorch: Low-Light Denoising - Alexis Baudron, Sony
2:10pm • Maximizing Training Throughput Using Torch.Compile and FSDP - Linsong Chu & Antoni Viros i Martin, IBM Research; Brian Vaughan, IBM
2:40pm • Running State-of-Art Gen AI Models on-Device with NPU Acceleration - Felix Baum, Qualcomm
3:10pm • TorchInductor CPU Backend Advancements: New Features and Performance Improvements - Jiong Gong & Leslie Fang, Intel
4:00pm • [HALIDE] A Halide Backend for TorchInductor - Jason Ansel, Meta
4:10pm • [MLIR] Enabling Composition of Kernels and Compilers - Jacques Pienaar, Google
4:20pm • [TRITON] Maximizing Kernel Development Productivity under Performance Constraints - Philip Tillet, OpenAI
4:30pm • [TVM] Universally Deploy Large-language Models via ML Compilation - Tianqi Chen, CMU & OctoAI
4:40pm • [MOJO] Lifting PT to New Heights with MAX and Mojo - Mikhail Zolotukhin, Modular
4:50pm • Together Goes Brrr: Threading Research & Production with Torch Compile - Pragaash Ponnusamy, together.ai
5:00pm • DL Compiler Panel Discussion - Philip Tillet, OpenAI; Jason Ansel, Meta; Jacques Pienaar, Google; Tianqi Chen, CMU & OctoAI; Mikhail Zolotukhin, Modular; Peng Wu, Meta
9:00am • Keynote: Welcome & Opening Remarks - Matt White, Executive Director, PyTorch Foundation
9:12am • Keynote: PyTorch Technical Deep Dive - Piotr Bialecki, NVIDIA; Peng Wu, Will Constable, Kartikay Khandelwal & Mengtao (Martin) Yuan, Meta
10:14am • Keynote: Open Language Models (OLMo): Accelerating the Science of Language Modeling - Hanna Hajishirzi, Senior Director NLP Research, Allen Institute for AI
10:30am • Keynote: Enabling Generative AI on the Edge - Cormac Brick, Principal Engineer, Google
1:15pm • Sponsored Keynote: The Lightning AI OSS Stack for Accelerating the AI Lifecycle - Luca Antiga, CTO, Lightning AI
1:20pm • Sponsored Keynote: Enabling AI Everywhere with PyTorch and Intel - Kismat Singh,VP of Engineering for AI Frameworks, Intel
1:30pm • Sponsored Keynote: From Containers to Cognition: Conducting the AI Orchestra - Taylor Dolezal, Head of Ecosystem, The Linux Foundation (CNCF)
1:35pm • Keynote Panel Discussion: Responsible AI - Kate Rooney, CNBC; Kush Varshney, IBM T. J. Watson Research Center; Sara Hooker, C4AI; Aleksander Madry, OpenAI; and Rishi Bommasani, Stanford University
11:10am • Meta Llama 3 and the Future of Responsible AI Development - Spencer Whitman & Vincent Gonquet, Meta
11:40am • Building Scientific Computing Infrastructure Software with the PyTorch Ecosystem - Bharath Ramsundar, Deep Forest Sciences
2:10pm • The Impact and Challenges of Open Source Generative Datasets and Models - Aaron Gokaslan, Cornell University
2:40pm • Lightning Talk: Beyond Zero: Eliminating Vulnerabilities in PyTorch Container Images - Patrick Smyth, Dan Fernandez & Srishti Hegde, Chainguard
3:10pm • Lightning Talk: PyTorch/XLA Auto-Sharding - Yeounoh Chung, Google
3:25pm • Lightning Talk: Introduction to Torch.Distributed.Pipelining - Howard Huang & Ke Wen, Meta
4:00pm • Lightning Talk: Debiasing the Data Lifecycle - Shailvi Wakhlu, Shailvi Ventures LLC
4:30pm • A Distributed Stateful Dataloader for Large-Scale Pretraining - Davis Wertheimer, IBM & Linsong Chu, IBM Research
5:00pm • Pushing the Performance Envelope: An Optimization Study for 3D Generative Modelling with PyTorch - Suvaditya Mukherjee & Shireen Chand, University of Southern California
10:50am • Lightning Talk: On-Device Profiling and Debugging with ExecuTorch - Olivia Liu & Vaun Puri, Meta
11:05am • Lightning Talk: LLMs on Edge with AI Accelerators - Chen Lai, Kimish Patel & Cemal Bilgin, Meta
11:20am • Sponsored Session: Torchchat: A Showcase of PyTorch LLM Ubiquity - Jack Khuu & Jesse White, Meta
11:50am • Lightning Talk: New Activation Checkpointing APIs in PyTorch - Jeffrey Wan & Horace He, Meta
12:00pm • Lightning Talk: FlexAttention - the Flexibility of PyTorch + the Performance of FlashAttention - Yanbo Liang & Horace He, Meta
12:10pm • Lightning Talk: Making the Most of Heterogeneous Memory Capacity Using PyTorch - Syed Ahmed, NVIDIA Corporation
2:15pm • Data-Dependent Shapes in PT2 - Edward Yang, Meta
2:45pm • Lightning Talk: What's New for PyTorch Developer Infrastructure - Sahan Paliskara & Catherine Lee, Meta
3:00pm • Lightning Talk: PyTorch Release Process - Andrey Talman, Meta
3:15pm • Torch.Compile for Autograd, DDP and FSDP - Will Feng , Chien-Chin Huang & Simon Fan, Meta
4:05pm • Startup Showcase
10:50am • The Rise of `Transformers` in the Growing PyTorch Ecosystem - Arthur Zucker, Hugging Face
11:20am • Training MoEs at Scale with PyTorch - Mihir Patel & Brian Chu, Databricks
11:50am • Lightning Talk: Empowering Developers: Tools and Resources for Running Generative AI on Arm CPUs - Pareena Verma, Arm
12:00pm • Lightning Talk: Optimized PyTorch Inference on aarch64 Linux CPUs - Sunita Nadampalli, Amazon (AWS)
12:10pm • Lightning Talk: AOTriton: Ahead of Time Triton Kernel Libraries on ROCm - Jeff Daily, AMD
2:15pm • vLLM: Easy, Fast, and Cheap LLM Serving for Everyone - Woosuk Kwon, UC Berkeley & Xiaoxuan Liu, UCB
2:45pm • Torchtitan: Large-Scale LLM Training Using Native PyTorch 3D Parallelism - Wanchao Liang, Meta & Linsong Chu, IBM Research
3:15pm • Slaying OOMs - Mark Saroufim & Jane Xu, Meta
4:05pm • Understanding the LLM Inference Workload - Mark Moyou, NVIDIA
4:35pm • Intel GPU in Upstream PyTorch: Expanding GPU Choices and Enhancing Backend Flexibility - Eikan Wang & Min Jean Cho, Intel
5:05pm • Implementing a Custom Torch.Compile Backend - A Case Study - Maanav Dalal & Yulong Wang, Microsoft
9:00am • Keynote: Welcome Back & Opening Remarks
9:07am • Keynote: Why You Should Think Twice Before Paying for an Evaluation Tool - Chip Huyen, VP of AI & OSS, Voltron Data
9:24am • Keynote: Navigating the Architectural Timeline of LLMs - Sebastian Raschka, Staff Research Engineer, Lightning AI
9:41am • Keynote: Building an Advanced Knowledge Assistant - Jerry Liu, Co-Founder & CEO, LlamaIndex
9:58am • Keynote: Ray: A Distributed Framework for Heterogeneous Computing - Ion Stoica, Professor, UC Berkeley
10:15am • Keynote: Community Awards
1:25pm • Sponsored Keynote: Accelerating AI: How AMD and PyTorch Drive Innovation with Seamless Day-0 Support and High Performance - Anush Elangovan, CVP Software Development, AMD
1:32pm • Sponsored Keynote: Optimizing AI Inference for Large Language Models - Mudhakar Srivatsa, Distinguished Engineer, IBM
1:40pm • Keynote Panel Discussion: Scaling & Benchmarking - Wei-Lin Chiang & Lisa Dunlap, UC Berkeley; James Bradbury, Anthropic; Tri Dao, together.ai; Aparna Ramani & Soumith Chintala, Meta
10:50am • Sponsored Session: Democratizing AI: Powering the Future with Arm’s Global Compute Ecosystem - Gian Marco Iodice, Arm
11:20am • Lightning Talk: Building and Supporting the Chinese PyTorch Community: Resources, Tutorials, and Engagement - Zong Zesheng, Huawei
11:35am • Lightning Talk: Distributing a Million Open Models in the Wild: Lessons Learned from the Hugging Face Hub - Omar Sanseviero, Hugging Face
11:50am • Lightning Talk: Implementing and Using Iterable Datasets: What Could Go Wrong? - Nicolas Hug, Meta
12:00pm • Lightning Talk: Fast, Scalable Distributed Training with StreamingDataset - Saaketh Narayan, Databricks
12:10pm • Lightning Talk: PyTorch-Wildlife: A Collaborative Deep Learning Framework for Conservation - Zhongqi Miao, Microsoft
2:15pm • Building PyTorch Computer Vision Algorithms for 100 Skin Shades - Emmanuel Acheampong, roboMUA
2:45pm • Blobs to Clips: Efficient End-to-End Video Data Loading - Andrew Ho & Ahmad Sharif, Meta
3:15pm • Sponsored Session: PyTorch Support by Google Enabling Performance from Cloud to Edge - Mark Sherwood & Shauheen Zahirazami, Google
4:05pm • Lightning Talk: Understanding and Optimizing PyTorch Models with Thunder - Luca Antiga, Lightning AI
4:20pm • Lightning Talk: d-Matrix LLM Compression Flow Based on Torch.Fx: Simplifying PTQ/QAT - Zifei Xu & Tristan Webb, d-Matrix Corporation
4:35pm • Unlocking the Enigma: Crafting Unbiased, Transparent, and Explainable Large Language Models - Rashmi Nagpal, Patchstack
5:05pm • The Ethical Implications of AI and the Environment: A Focus on Water - Amber Hasan, Ethical Tech AI & Senegal Tuklor Williams, Broken Pencil Pictures llc