SGLang-Diffusion serving framework for diffusion models to be unveiled at PyTorch Conference 2026
PyTorch · x · 2026-10-10
PyTorch announced that Yihao Wang and Kevin Mi of RadixArk will present SGLang-Diffusion at PyTorch Conference North America 2026 (San Jose, Oct 20-21). The framework targets high-performance serving of diffusion models for both large-scale offline generation and latency-sensitive real-time inference, with talks covering request scheduling, memory management, batching strategies, and execution-path optimizations built on PyTorch. The linked page also lists conference registration: $599 early bird / $799 standard / $999 late for attendees, and a flat $249 academic rate.
Related event: SGLang-Diffusion: High-Performance Serving Framework for Diffusion Models(2 posts)→
More from Infra
- llama.cpp only hits 8-10 t/s on a 4bit 27B while bitsandbytes + transformers manages 27 t/s — cephaloform · 2026-10-10
- Two H100 price indices show just 0.17 weekly correlation, clouding compute futures hedge — BenBajarin · 2026-10-10
- Chrome's new echo canceller halves voice agent word error rate, stops agents answering their own greeting — chadwallacehart · 2026-10-10
- Price war math: does cheaper AI tokens boost or drain compute investment? — dnlkwk · 2026-10-10
- New Paper 'Inference Auctions' Brings Market Mechanisms to LLM Inference Serving — nhaghtal · 2026-10-10
- Andrew Ng: one of my agents makes 5,000–10,000 web searches a day, data centers will need to scale far beyond current plans — DeepLearningAI · 2026-10-10