SGLang Summit 2026 by LMSYS

EVENT PAGE: sglang.io/summit

Two days with the people shaping frontier AI.

SGLang Summit 2026 brings the whole AI stack to one stage: the teams building next-generation chips, the labs training open and frontier models, the engineers running them in production, and the founders putting them to work.

Join 2,000+ engineers, researchers, builders, and leaders for two days of talks, deep dives, and workshops across frontier AI infrastructure on Nov 12-13, at Fort Mason, San Francisco 🌉

Meet the Speakers

Lip-Bu Tan (Intel) · Lilian Weng (OpenAI) · Soumith Chintala (Thinking Machines Lab) · Aravind Srinivas (Perplexity), and more

Highlights from Main Stage

Nov 12

• Opening Keynote: SGLang Ecosystem. What's next for SGLang and the ecosystem, spanning training and inference, broader hardware support, production infrastructure, and image & video generation.

• Open Models. The people building today’s leading open models on what they’ve learned the hard way: which bets paid off, which assumptions broke, and where the next generation is going.

• RSI (Recursive Self-Improvement). From models that help train their successors to systems that rewrite their own kernels, dive deep into the research, the training methods, the hard limits, and the questions no one has answered, or even asked, yet.

• AI Hardware. Perspectives on the next generation of AI hardware, covering GPUs, custom AI accelerators, system design, and the changing compute requirements of training and inference, with a candid conversation about squeezing more performance and efficiency out of every watt.

Nov 13

• AI Agent. Personalized intelligence, agent harnesses, and software factories: how agents are changing software development, everyday work, and scientific research. Find out what's working, what's breaking, and what caught us off guard.

• Neo Labs. Research perspectives from emerging AI labs, covering the questions guiding their work, their technical approaches, and new directions they are pursuing.

• AI Applications. Lessons from building AI products for consumers and enterprises, from product design and deployment to integration into everyday workflows.

• Physical Intelligence. Advances in robotics and world models, including learning, prediction, and action in physical environments, alongside alternative approaches to intelligence beyond LLMs.

Across Both Days

Explore the GPU Expo and exhibitor booths for live demos and conversations with the teams behind the products. Join technical breakouts and hands-on workshops, share your latest project on the Open Stage, and meet potential collaborators. Lounges, coffee chats, and community meetups offer space to exchange ideas and keep the conversations going between sessions.

Topics include models, reinforcement learning, AI hardware, cloud infrastructure, generative media, creator tools, interactive models, evaluation, local AI, AI applications, robotics & world models, production stack, and recommendation systems.

Plus a variety of community events. Stay tuned.

For ticket and sponsorship, see the full FAQ or contact us at summit@lmsys.org.