First SGLang Summit agenda revealed: Meta, NVIDIA, Perplexity to speak
bigdata · x · 2026-11-13
The first SGLang Summit 2026 runs November 12–13 at Fort Mason, San Francisco, under the theme "Open Foundations for the Next Era of AI," covering the full open AI stack from engine to hardware.
Agenda highlights:
- AI in Production: Cloudflare on large-scale deployment; LinkedIn on serving a small ads-ranking model at 10K+ QPS with scoring latency under 80 ms.
- Core Engine: Meta on bringing PyTorch to SGLang; Alibaba extending HiCache for long-running agents.
- Agentic Engineering: Perplexity's minimalist trainer for trillion-parameter models built with coding agents; Modal on agents developing and managing SGLang inference services.
- Training & Post-Training: Thinking Machines Lab and Meta on TorchTitan with Miles; Decagon on RL improving instruction following and cutting hallucinations.
- Hardware: NVIDIA on porting SGLang to Vera Rubin; Google on SGLang-TorchTPU for global-scale production inference.
- Real-Time AI & Media: SGLang Omni's interruptible conversations; HeyGen on GPU cost optimization; MiniMax on film AI workflows.
Related event: First SGLang Summit Opens in San Francisco with OpenAI, NVIDIA, Meta(2 posts)→
More from Infra
- AMD CEO Lisa Su, Jensen Huang's cousin, named Yahoo Finance's 2026 CEO of the Year — Polymarket · 2026-11-13
- Prime Intellect grew 100x on OpenRouter to 1T+ monthly tokens, cutting cost per token ~60% — willcb · 2026-11-13
- Compute is going fungible: Anthropic roadshow backs Ben Bajarin's hyperscaler thesis — BenBajarin · 2026-11-13
- Liquid AI's 74M-param LiquidCodec hits 361x realtime encoding on a single CPU thread — helloiamleonie · 2026-11-13
- AI Agent Traffic on Cloudflare Up 1,700% in a Year as Over Half of Web Traffic Turns Non-Human — PeterDiamandis · 2026-11-13
- One 8×B200 node serves 3,000 DeepSeek agents, 19x official capacity — james_y_zou · 2026-11-13