All talks from PyTorch Conference Europe 2026

0:53

1

What PyTorch Conference Europe 2026 Was Really Like – Official PyTorchCon EU Highlights | Paris

This talk reflects on PyTorch Conference Europe 2026, which took place on April 7–8 in Paris, France...

9:50

2

Lightning Talk: How DeepInverse Is Solving Imaging in Science and H... Andrew Wang & Minh Hai Nguyen

This talk by Andrew Wang from DeepInverse and Minh Hai Nguyen from Université de Toulouse explores h...

25:37

3

Why WideEP Inference Needs Data-Parallel-Aware Scheduling - Maroon Ayoub & Tyler Michael Smith

This talk, presented by Maroon Ayoub from IBM and Tyler Michael Smith from Red Hat, explores the cha...

19:17

4

Write Once, Run Everywhere with Pytorch Transformers - Pedro Cuenca, Hugging Face

This talk covers the integration of the Hugging Face transformers library, built on pure PyTorch, in...

21:08

5

Build PyTorch to Understand PyTorch - Vijay Janapa Reddi & Andrea Mattia Garavagno

This talk features Vijay Janapa Reddi from Harvard University and Andrea Mattia Garavagno from the U...

9:51

6

Lightning Talk: Why Your Forecasting Transformer Isn’t Working (And How To Fix It... Rosheen Naeem

This talk focuses on the challenges and solutions for improving forecasting transformers in Python,...

12:20

7

Lightning Talk: Bayesian Neural Networks With Variational Inference in PyTorch - Lars Heyen

This talk covers the implementation of Bayesian Neural Networks (BNNs) using Variational Inference i...

11:15

8

Lightning Talk: TerraKit: Standardising AI-Ready Geospatial Data... Rosie Lickorish & Romeo Kienzler

This talk introduces TerraKit, an open-source Python library designed to standardize the preparation...

10:24

9

Lightning Talk: ExecuTorch on Microcontrollers: Deploying PyTorch To... RJ Ascani & Matthias Cremon

This talk features RJ Ascani and Matthias Cremon from Meta as they discuss ExecuTorch, an extension...

28:38

10

Enabling State-of-the-art Asynchronous Execution in Torch.compile With CUDA Streams - Michael Lazos

This talk by Michael Lazos from Meta explores the enhancement of asynchronous execution in Torch.com...

22:13

11

The Token Slice: Implementing Preemptive Scheduling Via Chunked Decod... Maroon Ayoub & Kellen Swain

In this talk, Maroon Ayoub from IBM and Kellen Swain from Google present the concept of Chunked Deco...

10:06

12

Lightning Talk: Deep Learning in the Wild: Embedded PyTorch for... Taraqur Rahman & Owen O'Donnell

This talk focuses on the use of deep learning in wildlife conservation, specifically through the dep...

26:53

13

The Science and Practice of Open and Scalable LLM Evaluations - Grzegorz Chlebus, NVIDIA

This talk by Grzegorz Chlebus from NVIDIA delves into the science and practice of open and scalable...

24:58

14

Lights, Camera, Inference! Video Generation as a Service With VLLM-O... Ricardo Noriega & Doug Smith

This talk covers the exploration of video generation as a service using vLLM-Omni and the LTX-2 open...

20:31

15

Keynote: PyTorch Updates - Edward Yang, Research Engineer, Meta

This keynote by Edward Yang, a Research Engineer at Meta, covers the latest updates in PyTorch, a po...

24:01

16

Tour De Force: LLM Inference Optimization From Simple To Sophisticated - Christin Pohl, Microsoft

This talk by Christin Pohl from Microsoft focuses on optimizing large language model (LLM) inference...

9:57

17

Lightning Talk: Bringing Google’s Colossus to PyTorch: Rapid Stor... Ankita Luthra & Trinadh Kotturu

This talk examines the transition from compute to storage bottlenecks in PyTorch models as they scal...

24:27

18

Helion 1.0: A High-Level DSL for Performance Portable Kernels - Oguz Ulgen, Meta

This talk features Oguz Ulgen from Meta discussing Helion 1.0, a high-level domain-specific language...

24:53

19

torch.compile and Diffusers: A Hands-On Guide to Peak Performance - Sayak Paul, Hugging Face

This talk covers the integration of torch.compile with the Diffusers library to enhance the performa...

27:00

20

Optimizing PyTorch on CPU-GPU Coherent Platforms - Matthias Jouanneaux, Nvidia

This talk covers the optimization of PyTorch on Nvidia's GB200 coherent platform, highlighting the b...

10:23

21

Lightning Talk: Jigsaw: Domain and Tensor Parallelism for High-Resolution Inp... Deifilia Kieckhefen

This talk explores Jigsaw, a PyTorch library designed to enhance domain and tensor parallelism for h...

24:33

22

Why Classic IAM Collapses for Agents: Rethinking IAM for Agentic Systems - Parul Singh, Red Hat

This talk, presented by Parul Singh from Red Hat, explores the limitations of traditional Identity a...

33:07

23

Securing Agentic AI With PyTorch: Threat Modeling & LLM Red Teaming in Practice - Valeri Milke

This talk focuses on securing agentic AI systems built with PyTorch, highlighting the unique securit...

11:53

24

Keynote: Co-Evolution: How the Open Source Intelligence Stack Compounds - Mark Collier

This keynote presentation by Mark Collier, Executive Director of the PyTorch Foundation and General...

8:54

25

Lightning Talk: Implementing Single-Dim Strategies With Sharding Validator - Anshul Sinha, Meta

This talk, presented by Anshul Sinha from Meta, explores the implementation of Single-Dim Strategies...

12:13

26

Keynote: Community Led Open Source RL - Joe Spisak

This keynote features Joe Spisak, VP of Product and Head of Open Source at Reflection AI, discussing...

14:20

27

Lightning Talk: Flexible Deployment of PyTorch Models on MCU-Class... Robert Kalmar & Martin Pavella

This talk covers the efficient deployment of PyTorch models on MCU-class devices using ExecuTorch, a...

9:54

28

Lightning Talk: TorchJD: Jacobian Descent in PyTorch - Pierre Quinton & Valérian Rey

This talk covers Jacobian descent (JD), an advanced optimization technique that extends gradient des...

15:39

29

Lightning Talk: Cross-Region Model Serving: PyTorch Inference, Observability... Suraj Muraleedharan

This talk covers the challenges organizations face in deploying, monitoring, and operating PyTorch m...

11:41

30

Lightning Talk: Coding Agents for Compiler Construction: Beyond the... Reza Rahimi & Stefan Krassin

This talk focuses on coding agents as engineering tools for compiler construction, specifically with...

24:34

31

Optimizing Reinforcement Learning at Trillion-Parameter Scale - Songlin Jiang

This talk covers the implementation and optimization of reinforcement learning at the trillion-param...

23:12

32

Sponsored Session: TorchTPU: Expanding TPU Programmabil... Kat Ko, Claudio Basile & Jana van Greunen

This talk covers TorchTPU, a new initiative by Google aimed at expanding TPU programmability to PyTo...

12:48

33

Lightning Talk: Training Embedding Model Resiliently for Multimodal M... Huamin Chen & Haichen Zhang

This talk explores the training of embedding and classification models for efficient routing in Larg...

9:46

34

Lightning Talk: Ethical, Privacy and Sustainability Considerations in PyTorch S... Paula Mesa Macias

This talk by Paula Mesa Macias from Pau&Company examines the ethical, privacy, and sustainability co...

30:20

35

Brevitas Quantization Library - Pablo Monteagudo Lago, AMD

This talk covers the Brevitas Quantization Library, an open-source PyTorch resource developed by AMD...

4:45

36

Sponsored Keynote: From One Node to Distributed Training and Inference. How the PyTo... Ramine Roane

This keynote presentation by Ramine Roane, Corporate Vice President of AI Product Management and Eco...

24:23

37

Teaching PyTorch To Read Your Worst PDFs With Docling - Mingxuan Zhao, Peter Staar & Carol Chen

This talk focuses on the challenges of extracting clean, structured data from real-world documents,...

12:41

38

Lightning Talk: Running ExecuTorch Applications With Silicon Accelera... George Gekov & Aki Makkonen

This talk explores the efficient deployment of machine learning models on low-power embedded systems...

9:42

39

Lightning Talk: From Pretrained To Personal: Privacy-First... Daniel Holanda Noronha & Iswarya Alex

This talk explores the advancements in using PyTorch on AI PCs for meaningful model fine-tuning whil...

11:15

40

Keynote: The Unbearable Lightness of (Agentic) Evaluations - Besmira Nushi

This talk by Besmira Nushi, Senior Manager of AI Research at NVIDIA, explores the evolving disciplin...

26:01

41

Parameterized CUDA Graph Launch in PyTorch: CUDA Graphs Without the Pain - Daniel Galvez, NVIDIA

This talk focuses on the challenges of using CUDA Graphs in PyTorch, highlighting the potential bott...

25:23

42

On-Device LLM Inference on Android With ExecuTorch and Qualcomm QNN - Shivay Lamba & Kartikey Rawat

This talk covers the execution of large multimodal models like CLIP on Android devices using ExecuTo...

4:15

43

Sponsored Keynote: Any [ Agent | Model | Accelerator | Cloud ]... Maryam Tahhan & Nicolò Lucchesi

This talk features Maryam Tahhan and Nicolò Lucchesi from Red Hat, who discuss the advancements in o...

9:57

44

Lightning Talk: Combo Kernels: Horizontal Fusion Optimizat... Karthick Panner Selvam & Elias Ellison

This talk by Karthick Panner Selvam and Elias Ellison from Meta focuses on combo kernels, a compiler...

24:07

45

How To Write C++ Extensions in 2026 - Jane Xu, Meta & Mikayla Gawarecki, Meta

This talk covers the creation of C++ custom operation extensions for PyTorch in 2026, presented by J...

10:31

46

Lightning Talk: Inside VLLM's KV Offloading Connector: Async Memory Transfers for... Nicolò Lucchesi

This talk covers the KV Offloading Connector in vLLM 0.11.0, which addresses the challenge of recomp...

11:30

47

Lightning Talk: Accelerating On-Device ML Inference With ExecuTorch and Arm SME2 - Jason Zhu, Arm

This talk explores the challenges of achieving low-latency inference for complex on-device AI worklo...

23:25

48

From Responses To Trajectories: Multi-Turn and Multi-Environ... Kashif Rasul & Sergio Paniego Blanco

This talk covers the advancements in post-training large language models (LLMs) using reinforcement...

12:54

49

Lightning Talk: From Hugging Face To Handheld: Scaling LLM Deployment W... Cormac Brick & Weiyi Wang

This session explores the end-to-end journey of deploying custom PyTorch-based Open Source Large Lan...

20:51

50

Beyond the Theory: What Actually Breaks When You Scale Your Disaggregat... Ekin Karabulut & Ron Kahn

This talk focuses on the challenges of scaling disaggregated PyTorch models for inference in high-de...

20:16

51

Bringing PyTorch Monarch to AMD GPUs: Single-Controller Distributed Tra... Liz Li & Zachary Streeter

This talk covers the integration of PyTorch Monarch with AMD GPU technologies, focusing on single-co...

11:53

52

Lightning Talk: Why Logging Isn’t Enough: Making PyTorch Training Regressions Vi... Sahana Venkatesh

This talk presents an innovative approach to identifying training regressions in PyTorch, emphasizin...

10:24

53

Lightning Talk: Graph Based Pipeline Parallelism - Sanket Purandare, Meta & Simon Fan, Meta PyTorch

This talk discusses the challenges of pipeline parallelism in large models and how current PyTorch i...

9:38

54

Lightning Talk: Ball Tracking and Detection in Soccer Videos - Comparison of... Maciej Szymkowski

This talk focuses on ball tracking and detection in soccer videos, comparing Vision-Language Models...

11:48

55

Lightning Talk: Backpropagation-Free Optimization in PyTorch - Andrii Krutsylo

This talk, presented by Andrii Krutsylo from the Polish Academy of Sciences, explores backpropagatio...

13:54

56

Lightning Talk: Every Millisecond Counts: The Fine-tuning Journey of an Ultra-Eff... Pavel Macenauer

This talk, presented by Pavel Macenauer from NXP Semiconductors, delves into the optimization journe...

23:14

57

Bringing ExecuTorch To the Next Frontiers of Edge AI - Mergen Nachin, Meta

This talk features Mergen Nachin from Meta discussing the ongoing advancements of ExecuTorch, an on-...

14:04

58

Lightning Talk: Full-Stack PyTorch Robotics VLA: From Data To... Samet Akcay & Dmitriy Pastushenkov

This talk covers the transition of Vision-Language-Action models to production in robotics, addressi...

6:47

59

Lightning Talk: Slash LLM Cold-Start Times by Pre-distributing GPU... Billy McFall & Maryam Tahhan

This talk focuses on optimizing Large Language Model (LLM) deployments by addressing GPU cold-start...

10:27

60

Lightning Talk: Not All Tokens Are Equal: Semantic KV-Cache for Agen... Maroon Ayoub & Hyunkyun Moon

This talk, presented by Maroon Ayoub from IBM Research and Hyunkyun Moon from moreh, explores the co...

10:08

61

Lightning Talk: Debugging the Undebuggable: Introducing Torch.distributed.debug - Tristan Rice

This talk covers the complexities of debugging in distributed training using PyTorch and introduces...

10:23

62

Lightning Talk: KV-Cache Centric Inference: Building a State-Aware... Maroon Ayoub & Martin Hickey

This talk focuses on KV-cache centric inference, highlighting the significance of state management i...

9:21

63

Lightning Talk: FlexAttention + FlashAttention-4: Fast and Flexible - Driss Guessous, Meta

This talk features Driss Guessous from Meta, who discusses the integration of FlexAttention with Fla...

11:35

64

Lightning Talk: Beyond Generic Spans: Distributed Tracing for Actio... Sally O'Malley & Greg Pereira

This talk discusses the critical need for end-to-end observability in production large language mode...

10:01

65

Lightning Talk: Step-Aligned Telemetry for Distributed PyTorch Training (Time... Abhinav Srivastav

This talk covers the challenges of distributed PyTorch training, highlighting how misalignment betwe...

21:35

66

Optimizing Large MoE Inference on NVIDIA Blackwell: NVFP4, ADP, and DualPipe Strat... Julien Demouth

This talk focuses on optimizing large Mixture-of-Experts (MoE) inference using NVIDIA Blackwell’s fi...

7:26

67

Lightning Talk: Live Migration of PyTorch GPU Nodes From Azure To European Clouds - Mike Krom

This talk covers the live migration of PyTorch GPU nodes from Azure to European cloud providers, add...

14:03

68

Lightning Talk: Enabling the Audio Modality for Language Models - Eustache Le Bihan, Hugging Face

This talk features Eustache Le Bihan from Hugging Face, who discusses the integration of audio into...

24:02

69

Optimizing CPU LLM Inference in PyTorch: Lessons From VLLM - Crefeda Rodrigues & Fadi Arafeh

This talk covers optimizing CPU-based large language model inference using vLLM as a case study in t...

26:05

70

TorchStore: What We Learned Building Distributed Storage Sol... Lucas P, Danielle P, Allen W, Amir A

This talk explores the challenges and solutions in building distributed storage systems for Asynchro...

9:41

71

Lightning Talk: Scaling Recommendation Systems To 2K GPUs and Beyond - Zain Huda, Meta

This talk covers the advancements in recommendation system scalability at Meta, focusing on the impl...

15:55

72

De-mystifying PyTorch for ASICs: When (and Why) To Move Your Development To AI A... Alpha Romer Coma

This talk explores the transition of machine learning development from traditional GPUs to AI accele...

21:47

73

Model-Changing Transforms With Torch.compile - Thomas Viehmann, Lightning AI

This talk covers the capabilities of torch.compile, a tool designed to enhance the performance of Py...

24:32

74

PyTorch Symmetric Memory + NCCL Device APIs: A New Path Towards Multi-GP... Ke Wen & Sylvain Jeaugey

This talk covers the advancements in multi-GPU kernel development using PyTorch Symmetric Memory and...

10:20

75

Lightning Talk: Building a PyTorch‑native VLLM Plugin for IBM Spyre - Thomas Parnell & Thomas Ortner

This talk covers the redesign of the vLLM plugin for IBM Spyre, an AI accelerator employed in IBM Z...

25:24

76

PyTorch on RISC-V: From Cross-Compilation To Native CI - Ludovic Henry, Meta

This talk covers the efforts to port PyTorch natively to the RISC-V architecture, which is becoming...

5:12

77

Sponsored Keynote: Open Source Infrastructure for the AI Native Era - Jonathan Bryce

This keynote by Jonathan Bryce, Executive Director of the Cloud Native Computing Foundation, explore...

15:37

78

Keynote: Gemma 4: Compacting Intelligence for the Edge - Léonard Hussenot

This talk explores the philosophy and engineering behind Gemma 4, emphasizing that the future of AI...

15:08

79

Keynote: Stream Everything - Moving from Request input to Streaming input - Patrick von Platen

This talk features Patrick von Platen, a Research Engineer at Mistral AI, who presents insights on t...

10:21

80

Lightning Talk: Pluggable PyTorch LLM Inference Architecture With VLL... Yahav Biran & Maen Suleiman

This talk covers the evolution of PyTorch-based large language model (LLM) serving, focusing on the...

10:17

81

Lightning Talk: Faster Than SOTA Kernels in Torch.compile With Subgrap... Elias Ellison & Paul Zhang

This talk discusses how subgraph optimization and custom operator autotuning in torch.compile can ac...

21:29

82

From Gradients To Governance: Making PyTorch Lineage-Aware - Kateryna Romashko & Clodagh Walsh

This talk, presented by Kateryna Romashko and Clodagh Walsh from Red Hat, explores the necessity of...

11:28

83

Lightning Talk: Torch-Spyre: Compiling To a Multi-core Dataflow Accelerator... D. Grove & O. Tardieu

This talk covers the development of Torch-Spyre, an open source project designed to compile for the...

24:28

84

Portable High‑Performance LLM Serving: A Triton Backend for... Burkhard Ringlein & Jan van Lunteren

This talk covers the introduction of a Triton backend for vLLM, which is becoming the industry stand...

8:40

85

Lightning Talk: Bridging the Gap: Engineering Compliant... Muhammad Saqib Hussain & Mohaddisa Maryam

This talk features Muhammad Saqib Hussain and Mohaddisa Maryam from Neurosonic, focusing on creating...

10:24

86

Keynote: The Hub as Infrastructure. From Open PyTorch Models, to a Safe and Perfor... Lysandre Debut

This keynote features Lysandre Debut, Chief Open-Source Officer at Hugging Face, who discusses the c...

22:49

87

Seamless Integration: Custom Kernels in the Torch.compile Stack Wi... Kshiteej K, Masaki K & Pawel G

This talk focuses on the integration of custom kernels within the Torch.compile stack to enhance hig...

10:23

88

Keynote: PyTorch CTO - Matt White, Global CTO of AI, Linux Foundation

In this keynote, Matt White, Global CTO of AI and CTO at PyTorch Foundation, delivers an insightful...

25:59

89

Deploying PyTorch Models To the Browser and Beyond With Transformers.js - Joshua Lochner

This talk covers the engineering roadmap for running Hugging Face Transformers locally in web browse...

9:53

90

Lightning Talk: Building AI That Ops Teams Actually Trust - Robert King

This talk focuses on building AI systems that operations teams can trust, addressing the skepticism...

12:06

91

Lightning Talk: Distributed AI Without the Infrastructure Tax - Yahav Biran & Maen Suleiman

This talk addresses the challenges of running distributed AI workloads in production, focusing on pa...

25:24

92

Fp8 Training From Hopper To Blackwell - Luca Wehrstedt, Meta

This talk by Luca Wehrstedt from Meta explores the evolution of training with low-precision float8 d...

19:32

93

DualPipe from Scratch: Implementing DeepSeek's 5D Parallelism in PyTorch - Dev Jadhav, ING Bank

This talk covers the implementation of DeepSeek's 5D parallelism and DualPipe in PyTorch, addressing...

18:38

94

Building Trust for Users and Regulators Alike: A Cost-Efficient PyTorch Pat... Raja Gopal Hari Vijay

This talk explores the concept of Compliance-as-Code, focusing on how it integrates regulatory contr...

10:40

95

Keynote: vLLM & Ray Updates - Tyler Michael Smith & Artur Niederfahrenhorst

This keynote session features Tyler Michael Smith, Chief Architect of Inference Engineering at Red H...

19:04

96

Beyond JSON-RPC: Scaling Model Context Protocols With gRPC in the Py... Ashesh Vidyut & Madhav Bissa

This talk by Ashesh Vidyut and Madhav Bissa from Google addresses the limitations of Model Context P...

7:45

97

Lightning Talk: Achieving SOTA GEMM Performance: A CuTeDSL Backend for PyTorch Induc... Nikhil Patel

This talk covers the development of a new CuTeDSL backend for PyTorch Inductor that aims to achieve...

21:33

98

Bridging the Hardware Gap With Code Harnesses on the Hugging Face Kernels Hub - Ben Burtenshaw

This talk covers the development of code harnesses for standardizing kernel writing in the context o...

21:15

99

Sponsored Session: Fault-Tolerant Training: How We Build Rel... Cyril Konkratenko & Maurits de Groot

This talk covers fault-tolerant training for large-scale distributed AI workloads, highlighting Nebi...

12:40

100

Accelerating Complex-Valued Tensors With Torch.compile - Hameer Abbasi, OpenTeams Inc.

This talk covers the advancements in supporting complex-valued tensors with Torch.compile, presented...

14:01

101

Lightning Talk: Accelerating PyTorch Models With Torch.compile's C++ Wrapper Mode - Bin Bao, Meta

This talk introduces torch.compile's C++ wrapper mode, a feature that significantly reduces CPU over...

10:34

102

Lightning Talk: Trinity Large - Torchtitan on 2000+ B300s - Matej Sirovatka, Prime Intellect

This talk covers the use of torchtitan for scaling the training of ultra-sparse mixture-of-experts m...

11:54

103

Lightning Talk: Monarch: An API To Your Supercomputer - Marius Eriksen, Meta

This talk covers Monarch, a distributed programming framework for PyTorch that simplifies supercompu...