PyTorch

The 2026 PyTorch Docathon is a community event where contributors help improve the tutorials, guides, examples, and documentation that support the PyTorch ecosystem. Running May 5 to May 19, the Docathon includes beginner-friendly issues, skill-based tasks, and community support for both new and experienced contributors. It’s a practical way to contribute to PyTorch while helping make the project easier to learn and use. RSVP for updates and event details: https://lnkd.in/gXVN2TyM #PyTorch #PyTorchDocathon #OpenSourceAI

IBM Research introduced the RITS Platform to provide its research community with shared access to model inference endpoints. vLLM is at the heart of the RITS Platform and has been critical to democratizing access to the latest LLMs as they release. “The vLLM community is vibrant and responsive, and with collaborative expertise, we are able to do great things both upstream and internally by leveraging and contributing to this groundbreaking project. vLLM has been critical to democratizing access to our research community to the latest and greatest LLMs as they release.” – Priya Nagpurkar, Vice President, AI Platform, IBM Research Read our latest case study: 🔗 https://lnkd.in/gVjCnBDm #vLLM #PyTorch #OpenSourceAI

Mark Collier, Executive Director of PyTorch Foundation and General Manager of AI & Infrastructure at The Linux Foundation, opened PyTorch Conference Europe 2026 with a keynote on how the open source intelligence stack advances through coordination across hardware, software, and open source communities. He shared updates on our growth, our expanding set of hosted projects, and why open collaboration remains essential to building AI infrastructure at global scale. 🎥 Hear from Mark: https://lnkd.in/gHmMTHRw #PyTorchCon #PyTorch #vLLM #DeepSpeed #Ray #Helion #Safetensors #ExecuTorch #OpenSourceAI

What to streamline the process of packaging your apps for AI accelerated computing? Michael Kennedy of Talk Python Training interviews a coalition from NVIDIA, Astral, and QuantSight who have been working on Wheel Next: A set of PEPs that let packages declare what hardware they need and let installers like uv pick the right build automatically. Listen to the PodCast – ‘Wheel Next + Packaging PEPs’: https://lnkd.in/gnxvNdM6 #PyTorch #OpenSourceAI #AI #Inference #Innovation

llm-d published a new post on KServe + llm-d + vLLM for production LLM inference on Kubernetes. Authors from Red Hat and Tesla describe how the stack addressed routing, customization, and day-2 operational challenges, citing 3x higher output tokens/s and 2x lower TTFT in one deployment after enabling prefix-cache aware routing. By Yuan Tang, Scott Cabrinha, Robert Shaw, and Sai Krishna Cloud Native Computing Foundation (CNCF) 🔗 https://lnkd.in/gSSZNdVp #vLLM #KServe #Kubernetes #LLMOps #OpenSource

Join the global PyTorch community in the heart of Silicon Valley! 🚀 PyTorch Conference North America is coming to San Jose, CA, this October 20-21. This is the premier destination for developers, researchers, and innovators to collaborate on the future of AI. 🎟️ Register by July 31 to secure early bird rates and save $400! 🔗 Learn more and register: https://bit.ly/4sh3DSw #PyTorchCon #PyTorch #PyTorchFoundation #FutureOfAI #AI #GenAI #MachineLearning #ML #DeepLearning #OpenSource #OpenSourceSoftware #OpenSourceDevelopment #OpenSourceCommunity #OSS #LinuxFoundation #events #linux

PyTorch Conference Europe 2026 was one for the books. 2 days in Paris building the future of open source AI together. 100+ session recordings are now live. View the full #PyTorchCon EU playlist: https://lnkd.in/eGkjnSvW

Want to Integrate Physical AI Capabilities into Existing Apps? By stripping away the platform plumbing, Isaac Lab can leverage tensorized data exchange for direct, high-speed access to simulation states via GPU buffers (e.g., positions/velocities as PyTorch tensors without host copies). You can interact using zero-copy transfer data exchange directly alongside popular frameworks like NumPy, PyTorch, and Warp. 🔗 Read the full NVIDIA developer blog: https://lnkd.in/gxPawQFY #PyTorch #OpenSourceAI #AI #Inference #Innovation

The 2026 PyTorch Docathon will take place May 5 to May 19, a hackathon-style community event focused on improving PyTorch documentation. Whether you're new to open source or an experienced contributor, the Docathon offers opportunities to improve tutorials, guides, examples, and website content. Many issues are beginner-friendly, with tasks labeled by skill level and support available through the PyTorch Discord. Contributors can make an immediate impact while learning more about PyTorch modules, tutorials, and workflows. Schedule: May 5 Kickoff and Q&A, 10:00 AM PT May 6 to May 15 Submissions and feedback May 16 to May 18 Final reviews May 20 Winner announcements RSVP to receive updates, participation details, and event instructions. Learn more: https://lnkd.in/gXVN2TyM #PyTorch #PyTorchDocathon #OpenSourceAI

Meta teams pursue aggressive ROI goals, requiring substantial capacity reductions for model training and serving to meet financial targets. This is especially challenging for large-scale training jobs, which involve more data, GPUs, and advanced modeling techniques—resulting in higher initialization costs as models grow. To improve efficiency, Meta uses Effective Training Time (ETT%) to measure the proportion of end-to-end (E2E) wall time spent on productive training, factoring in overheads like initialization, restarts, checkpoint delays, and failures. Since 2024, teams have launched initiatives to minimize training job overhead. This blog reviews key focus areas, progress made, and next steps: https://lnkd.in/gq2bMnSP #PyTorch #OpenSourceAI #ETT #Optimization