iiinova — where your digital self keeps evolving

PyTorch’s latest blog covers Jagged Flash Attention on NVIDIA B200 with TLX. On GEM’s jagged shapes, the TLX kernel is about 13% faster on the forward pass and about 50% faster on the backward pass than the May 2026 version of FA4. bit.ly/4dhV84P

000

PyTorch @pytorch.org · 30/09/2026

In his talk at PyTorch Conference North America 2026, Suvaditya Mukherjee from Hugging Face will share guidance on how to use the PyTorch Profiler on your models to extract maximum performance efficiency. Join us in San Jose, Oct 20-21: hubs.la/Q04v4SL60 #PyTorchCon

000

PyTorch @pytorch.org · 30/09/2026

Ray has become critical across the full AI lifecycle, from data curation to production serving and reinforcement learning. Read the full guide and map out your schedule: bit.ly/4ysjCku Register for PyTorch Conference North America: bit.ly/464FTbp

010

PyTorch @pytorch.org · 30/09/2026

In a new community blog, contributors from IBM and Red Hat share how they built Torch Spyre on top of PyTorch's Cross-Repository CI Relay (CRCR) to tackle compatibility testing at scale. Read the full technical deep dive here: bit.ly/46XkjGi

011

PyTorch @pytorch.org · 29/09/2026

How does the PyTorch Ambassador Program work, and how can you position yourself to become one? At #PyTorchCon North America 2026, Sahdev Zala (IBM) will cover what ambassadors have built, challenges they’ve faced, and ways to engage more deeply. hubs.la/Q04v4SL60

000

PyTorch @pytorch.org · 29/09/2026

vLLM is a high-throughput, memory-efficient inference and serving engine for LLMs. At #PyTorchCon North America, Simon Mo’s keynote and vLLM technical sessions cover KV cache management, disaggregated serving, attention, and more. hubs.la/Q04vK-VQ0

000

PyTorch @pytorch.org · 29/09/2026

Communication overhead remains one of the largest bottlenecks in scaling distributed training. Join us at PyTorch Conference North America in San Jose on October 20-21 to explore the future of distributed training infrastructure: hubs.la/Q04v4SL60 #PyTorchCon

013

PyTorch @pytorch.org · 28/09/2026

In her upcoming session on Hardware-Aware AI at #PyTorchCon NA 2026, Kavya Sri Chennoju from Arm will demonstrate how PyTorch, ExecuTorch, and Arm Device Connect work together to deploy AI agents safely onto edge hardware. Join us in San Jose, October 20-21: hubs.la/Q04v4SL60

000

PyTorch @pytorch.org · 25/09/2026

Debugging LLM training in production is notoriously challenging. Join us at PyTorch Conference North America in San Jose, October 20 to 21, to learn how bitwise alignment delivers faster, more precise LLM training debugging: hubs.la/Q04v4SL60 #PyTorchCon

000

PyTorch @pytorch.org · 25/09/2026

At PyTorch Conference North America, Maajid Khan from Fujitsu Research India will explore efficient MoE LLM inference using vLLM and OpenVINO, sharing strategies for running these models effectively on ARM architectures. Join us in San Jose, October 20-21: hubs.la/Q04v4SL60 #PyTorchCon

000

PyTorch @pytorch.org · 24/09/2026

To support developers in deepening their technical skills, the PyTorch Foundation is running a dedicated Introduction Track and a full-day PyTorch Associate Training session at PyTorch Conference North America in San Jose, California Read our latest blog to learn more: bit.ly/4dE9899

012

PyTorch @pytorch.org · 24/09/2026

At #PyTorchCon North America 2026, Liz Li (AMD) and Shekhar Pandey (AMD) will discuss scalable MXFP8 pretraining on MI355X GPUs, including TorchAO kernel optimization and TorchTitan integration for an end-to-end upstream PyTorch training stack. hubs.la/Q04v4SL60

000

PyTorch @pytorch.org · 23/09/2026

At #PyTorchCon North America 2026, Alessandro Sangiorgi, Senior Software Engineer (Red Hat), will show how warm starts let Helion autotuning reuse cached configurations for faster iteration instead of repeated cold searches. hubs.la/Q04v4SL60

011

PyTorch @pytorch.org · 23/09/2026

The PyTorch Foundation OSPO & Academic Outreach Working Group is seeking academic open source projects for a Day 0 workshop at PyTorchCon NA 2026. Submit by Sept. 30: pytorch.org/blog/from-research-proj…

011

PyTorch @pytorch.org · 23/09/2026

Following a great community gathering at Graphcore's London office, momentum is building for PyTorch Conference North America 2026. Register now to save your spot: hubs.la/Q04v4SL60 #PyTorchCon

010

PyTorch @pytorch.org · 22/09/2026

In our latest blog, contributors from IBM, Meta, and Hugging Face introduce hardware-agnostic layers designed to balance frontier performance with portability, helping ensure vLLM continues to meet the needs of the broader open-source ecosystem. Read the blog to learn more: bit.ly/4xEH7Wk

010

PyTorch @pytorch.org · 21/09/2026

TinyTorch is a free, open source curriculum where you build a working machine learning framework from scratch, tensors through transformers, using PyTorch’s own API in pure Python. Read the full technical breakdown here: bit.ly/4ArNTRO

000

PyTorch @pytorch.org · 21/09/2026

At PyTorch Conference North America 2026, Valerian Rey from Simplex Labs will break down multi-objective optimization. Join us in San Jose, October 20-21 to learn how to streamline your pipeline: hubs.la/Q04v4SL60 #PyTorchCon

000

PyTorch @pytorch.org · 20/09/2026

Connect with the open source AI community at PyTorch Conference North America in San Jose on October 20-21 to discuss open source runtime performance, hardware optimization, and testing frameworks. Register for PyTorch Conference America today: hubs.la/Q04v4SL60 #PyTorchCon

152

PyTorch @pytorch.org · 19/09/2026

Join us in San Jose on October 20 and 21 to learn how CRCR streamlines testing workflows across the ecosystem. Register today: hubs.la/Q04v4SL60 #PyTorchCon

000

PyTorch @pytorch.org · 18/09/2026

At PyTorch Conference North America, explore how new optimizers like Muon and TorchJD work alongside tools like fastsafetensors and LMCache to drive multi-loss efficiency and faster LLM serving. Read the Open Research, Tooling & Optimization Sessions Guide: bit.ly/4gXxn4r #PyTorchCon

041

PyTorch @pytorch.org · 18/09/2026

The first-ever PyTorch Day Japan comes to Tokyo Dec. 10, co-hosted by PyTorch Foundation, Hugging Face, IBM, and Mitsubishi Electric. Explore PyTorch, vLLM, DeepSpeed, Ray, Helion, and Safetensors. CFP open through Sept. 27. pytorch.org/blog/pytorch-day-japan-…

000

PyTorch @pytorch.org · 16/09/2026

PyTorch Conference is the open source AI community's town square. Where what's next gets decided. Register now to join the community in San Jose, October 20-21: bit.ly/464FTbp Read the Open Research, Tooling & Optimization Sessions Guide here: bit.ly/4gXxn4r #PyTorchCon

020

PyTorch @pytorch.org · 16/09/2026

Meta Engineering extended FlashAttention-4 with MXFP8 support for NVIDIA Blackwell. LP FA4 reaches 2.85 PFLOP/s forward and 2 PFLOP/s backward, with up to 1.30× end-to-end speedup. pytorch.org/blog/low-precision-flas…

000

PyTorch @pytorch.org · 11/09/2026

Designed for early-stage practitioners, the PyTorch Certified Associate (PTCA) program helps you demonstrate foundational ability across model design, training, and deployment while building credibility across the open source AI ecosystem. Learn more and enroll here: bit.ly/4do6qEx

001

PyTorch @pytorch.org · 11/09/2026

The HuggingFace Kernels project now has Helion support. This blog walks through how to build, autotune, and ship performant and portable Helion kernels via the Hugging Face Kernels project, allowing users to consume these kernels seamlessly. Learn more here: bit.ly/4d9UOF6

020

PyTorch @pytorch.org · 11/09/2026

At PyTorch Conference North America 2026, Nicolò Lucchesi, Research Engineer at Mistral AI and vLLM maintainer, will talk about how disaggregated serving in vLLM has evolved to support the latest generation of hybrid models. Join us in San Jose on October 20-21: hubs.la/Q04v4SL60

000

PyTorch @pytorch.org · 11/09/2026

The PyTorch Korea User Group will host its inaugural offline conference on Saturday, November 21, 2026, hosted at AWS Korea in Centerfield East, Seoul. The deadline to submit proposals is Sunday, September 13, 2026, at 23:59 KST. Submit your talk today: bit.ly/4iVlME7

110

PyTorch @pytorch.org · 10/09/2026

Join Thomas Cottenier from Arm this October 20-21. Thomas will present how AI agents can synthesize customized, target-specific runtimes using PyTorch components like torch.export, ExecuTorch, and torchao. Register now for PyTorch Conference North America: hubs.la/Q04v4SL60 #PyTorchCon

030

PyTorch @pytorch.org · 10/09/2026

KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China 2026 brought the open source AI stack together in Shanghai. Read the recap: pytorch.org/blog/pytorch-conference… #PyTorchCon #KubeCon #CloudNativeCon #OpenInfraSummit

000

PyTorch @pytorch.org · 09/09/2026

At PyTorch Conference North America, Ricardo Noriega, Tech Lead for the vLLM Omni team and Alex Brooks, Principal Machine Learning Engineer at Red Hat, will share practical technical strategies for optimizing multistage AI workloads. Register today: hubs.la/Q04v4SL60 #PyTorchCon

011

PyTorch @pytorch.org · 08/09/2026

In this video, Arun Bhandari, PyTorch Foundation Ambassador and leader of PyTorch Community Nepal, shares why he is excited to connect with the community and explore the latest developments in PyTorch. Join us in San Jose, CA this October 20-21: hubs.la/Q04v88dJ0 #PyTorchCon

000

PyTorch @pytorch.org · 08/09/2026

Alibaba Cloud, Cambricon, and Ant Group join the PyTorch Foundation. At #PyTorchCon China, they join Huawei to advance the open AI stack. pytorch.org/blog/alibaba-cloud-ant-…

000

PyTorch @pytorch.org · 07/09/2026

At PyTorch Conference 2026, Lu Fang, Research Scientist, and Ilina Mitra, Engineering Manager at Meta, will present a deep dive on building high-performance recommendation inference systems. Register for PyTorch Conference North America today: hubs.la/Q04v4SL60 #PyTorchCon

010

PyTorch @pytorch.org · 07/09/2026

Connecting researchers, engineers, and open-source contributors in Bengaluru, Red Hat and Hugging Face hosted an evening focused on the infrastructure, abstractions, and systems shaping the open ML stack. Check out the full technical recap and session takeaways here: bit.ly/4xRUTpl

000

PyTorch @pytorch.org · 06/09/2026

At PyTorch Conference North America 2026, Jiho Kim and Junghwan Park present a poster on growing PyTorch Korea through translation, mentorship, and regional events. Register here to join the open source AI community in San Jose, October 20-21, 2026: hubs.la/Q04v4SL60 #PyTorchCon

010

PyTorch @pytorch.org · 05/09/2026

Join us at PyTorch Conference North America 2026 (Oct 20–21, San Jose) for deep dives into kernel engineering and performance optimization across GPUs, TPUs, NPUs, and custom ASICs. Register here: bit.ly/464FTbp Read the Sessions guide: bit.ly/4yoK5z2 #PyTorchCon

011

PyTorch @pytorch.org · 05/09/2026

At PyTorch Conference North America 2026 (October 20–21 in San Jose, CA), Avik Chaudhuri from Meta will break down how PyreFly - a static type checker for Python - enables end-to-end static shape coverage for PyTorch models. Register for #PyTorchCon North America today: bit.ly/4qStJvL

001

PyTorch @pytorch.org · 04/09/2026

PyTorch Conference North America 2026 brings the core developer community together this October 20-21 in San Jose. #PyTorchCon Last chance to register for standard ticket here: bit.ly/464FTbp Read the Hardware Acceleration & Compute Infrastructure Sessions guide: bit.ly/4yoK5z2

021

PyTorch @pytorch.org · 03/09/2026

At PyTorch Conference North America 2026, connect with people working across PyTorch, vLLM, DeepSpeed, Ray, Helion, Safetensors, and the AI stack. At #PyTorchCon NA 2025, Richard Liaw spoke about “how important Ray can be for the stack.” Register by Sept. 4: hubs.la/Q04tDf0j0 #PyTorchCon

021

PyTorch @pytorch.org · 03/09/2026

Discover how open source agentic search and hardware-guided workflows are unlocking massive speedups across GPUs and TPUs at PyTorch Conference North America (Oct 20–21 in San Jose). Register by September 4: bit.ly/464FTbp Read the Agentic AI guide: bit.ly/4xL2vK4 #PyTorchCon

120

PyTorch @pytorch.org · 02/09/2026

PyTorch 2.14 brings CuTeDSL-generated CUTLASS kernels to Inductor through NVGEMM, a new nccl2 backend, first-class c10d fault tolerance, native linear algebra on Apple Silicon, and declarative dynamic shapes via @dynamic_spec. Read more: bit.ly/4yC3v3B Live Q&A: bit.ly/4x4KtBG

000

PyTorch @pytorch.org · 02/09/2026

We’re bringing together open source leaders to tackle some of the biggest technical challenges in agentic AI at PyTorch Conference North America (Oct 20–21 in San Jose). Register by September 4: bit.ly/464FTbp Read the Agentic AI and Next-Gen Sessions guide: bit.ly/4xL2vK4

031

PyTorch @pytorch.org · 02/09/2026

PyTorch 2.14 adds NVGEMM for Inductor, a new nccl2 backend, fault-tolerant c10d reconfiguration, native linear algebra on Apple Silicon + dynamic shapes. Live Q&A Sept. 17: streamyard.com/watch/E7iTfnUvpvTW Learn more at #PyTorchCon Oct. 20–21: hubs.la/Q04tDf0j0

021

PyTorch @pytorch.org · 01/09/2026

Meet the people behind the open source tools we use every day. PyTorch Foundation Ambassador, Abdulsalam Bande, is looking forward to connecting with the wider PyTorch Community at PyTorch Conference North America this October 2026. Register for PyTorchCon NA: bit.ly/4qN2DGt

021

PyTorch @pytorch.org · 31/08/2026

NVIDIA's Anjulie Agrusa, Ryan Spring, and Bruce Zitelli will show how the latest NVFP4 pretraining recipes can accelerate large-scale LLM training while closing the quality gap to BF16 at #PyTorchCon. Register by September 4 to save: hubs.la/Q04tDf0j0

020

PyTorch @pytorch.org · 31/08/2026

vLLM sessions at #PyTorchCon NA 2026 cover kernel and inference optimization, PyTorch release validation and compatibility, and broader serving workflows. Topics include fastsafetensors, quantization, dynamic shapes, VLM serving, storage, and contributor onboarding. hubs.la/Q04tDf0j0

021

PyTorch @pytorch.org · 30/08/2026

vLLM at #PyTorchCon NA 2026 covers attention, KV cache management, disaggregated serving, expert parallelism, and hardware portability. Register by Sept 4: hubs.la/Q04tDf0j0 Guide: pytorch.org/blog/vllm-sessions-at-p…

030

PyTorch @pytorch.org · 29/08/2026

At #PyTorchCon NA 2025, Christian Jacobi (IBM) spoke about the innovation ahead in deploying AI across core IT and business processes at scale with enterprise qualities of service. #PyTorchCon North America returns Oct 20–21. Register by Sept 4: hubs.la/Q04tDf0j0

011

PyTorch @pytorch.org · 28/08/2026

At PyTorch Conference North America 2026, hear from the people working across PyTorch Foundation projects. Simon Mo (vLLM, Inferact) says vLLM’s goal is to become “the easiest to use and most efficient inference engine.” #PyTorchCon Register by Sept 4: hubs.la/Q04tDf0j0

020