PyTorch

In this clip from his PyTorch Conference Europe 2026 keynote, Patrick von Platen (Mistral) discusses why real-world machine interaction requires models that can take continuous input and produce continuous output. Using live transcription as an example, he explains how streaming architectures differ from traditional speech recognition approaches that process larger chunks of audio at once. 🔗 Watch the full keynote: https://lnkd.in/e6zviXyW #PyTorchCon

The inaugural PyTorch Meetup Singapore brought together engineers, researchers, and community builders to talk about everything from vLLM project updates to the broader question of sovereign intelligence. Read the full technical recap and find presentation slides in our latest blog: https://bit.ly/4vdcPJU Sudhir Dharanendraiah

🚀 The co-located event lineup just got bigger. We're excited to welcome OSPOlogy + OSPO Summit China to KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China in Shanghai on September 7. This dedicated event brings together OSPO leaders, open source strategists, maintainers, and community builders to explore the people, processes, and practices behind successful open source programs. 🎤 Call for Proposals closes July 12 💰 Add it to your event pass by June 30 for only ¥70 (USD$10) before rates increase Secure your spot and submit your talk: https://bit.ly/4uPnslb #OspoSummit #KubeCon #CloudNativeCon #OpenInfraSummit #PyTorchCon #Kubernetes #K8s #CloudNative #OpenInfra #PyTorch #PyTorchFoundation #FutureOfAI #AI #GenAI #MachineLearning #ML #DeepLearning #OpenSource #OpenSourceSoftware #OpenSourceDevelopment #OpenSourceCommunity #OSS #CNCF #CloudNativeComputingFoundation #events #linux #software

Enable smarter, longer-thinking agents Scale agentic AI and reinforcement learning by shortening CPU execution time, increasing task throughput, and improving overall AI factory output. The NVIDIA custom Olympus core in the NVIDIA Vera CPU uses a neural branch predictor to reduce stalls in branch-heavy code. Combined with other prediction mechanisms, it can sustain two taken branches per cycle with zero penalty, maintaining throughput for deep software stacks such as PyTorch, graph workloads, and scripting engines. Read the complete blog post: https://bit.ly/4vz8oZA

Got a project, experiment, workflow, or research result worth sharing? The Poster Sessions at PyTorch Conference North America offer an opportunity to connect directly with the PyTorch community and showcase your work in an interactive setting. Join us October 20-21 in San Jose, CA, and present your ideas to developers, researchers, engineers, and AI practitioners from across the ecosystem. 📅 Poster CFP deadline: July 26 Submit your proposal: https://bit.ly/41k9TOg Planning to attend? Register today: https://bit.ly/4sh3DSw #PyTorchCon #PyTorch #PyTorchFoundation #FutureOfAI #AI #GenAI #MachineLearning #ML #DeepLearning #OpenSource #OpenSourceSoftware #OpenSourceDevelopment #OpenSourceCommunity #OSS #LinuxFoundation #events #linux #CallForProposals #CallForPapers #CFP #CallForSpeakers

Helion is a PyTorch-native hardware agnostic kernel DSL designed for writing high-performance kernels using a tile-programming model. In our latest blog post, Xiaohong Chen (Red Hat) and Yanan Cao (PyTorch, Meta) explored integrating Helion kernels with vLLM and evaluated their performance across a range of LLM inference workloads, showing impressive performance speedups thanks to Helion’s Ahead-of-Time autotuning and fine-grained runtime dispatching. Let’s dive into kernel fusion, framework design, and end-to-end benchmarking, showing how Helion can be easily integrated into vLLM and bring significant benefits. Read the full technical deep dive 👉 https://lnkd.in/ezSKU2cX

Are you ready for the next evolution of open source? 🚀 KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China is the premier destination for practitioners in Shanghai this September 8-9. From large-scale cloud orchestration to cutting-edge AI training techniques, this event offers unparalleled technical depth. 🗓️ Important Date: Secure your registration by July 28 to save ¥710 RMB (USD$100). Join 1,000+ innovators at the Shanghai International Convention Center. 🔗 Details and registration: https://bit.ly/4cduhWr #KubeCon #CloudNativeCon #OpenInfraSummit #PyTorchCon #Kubernetes #K8s #CloudNative #OpenInfra #PyTorch #PyTorchFoundation #FutureOfAI #AI #GenAI #MachineLearning #ML #DeepLearning #OpenSource #OpenSourceSoftware #OpenSourceDevelopment #OpenSourceCommunity #OSS #CNCF #CloudNativeComputingFoundation #events #linux #software

perf packed release: safetensors 0.8.0 is out ⚡️ Main takeaways: - direct copy into metal MTLBuffers + dlpack for 0-copy hand-off to target framework (only torch for now) -> 2-3x perf improvement + fixes OOMs loading models that are around the limit of unified memory on macOS when loading with transformers - GIL-free serialization, enabling multi-threaded saves from Python -> 1.2x to 2x faster for single files, but you can expect more improvement when saving multiple files in parallel! Check the release notes for the full list of improvements!

AI agents are changing how you interact with your PC. NVIDIA and Microsoft are teaming up to enable the next generation of developers to build on-device agents on the Windows platform, with easier setup, native security, and integration with the apps and tools developers already use. This includes optimizing the ComfyUI implementation in PyTorch, enabling developers to run larger models with better performance. Read the complete blog post: https://bit.ly/3QvPgN9

The open source AI ecosystem moves fast. PyTorch Conference North America is where the community comes together to share what’s working, what’s next, and what’s being built right now. Join researchers, developers, engineers, and AI leaders in San Jose, California, October 20-21 for two days of technical sessions, keynotes, networking, and community collaboration. 💰 Register by July 31 and save $400. 📌 Have research, a project, or a unique use case to share? The Poster Session Call for Proposals is open through 26 July. Register: https://bit.ly/4sh3DSw #PyTorchCon #PyTorch #PyTorchFoundation #FutureOfAI #AI #GenAI #MachineLearning #ML #DeepLearning #OpenSource #OpenSourceSoftware #OpenSourceDevelopment #OpenSourceCommunity #OSS #LinuxFoundation #events #linux #CallForProposals #CallForPapers #CFP #CallForSpeakers