SGLang's 3-year evolution: from RadixAttention to inference engine behind 330+ orgs
FinanceYF5 · x · 2026-11-15
A retrospective video traces SGLang's journey: from the RadixAttention paper (Jan 2024) to v1.0 and the SGLang Consortium (Nov 2026), hitting twenty milestones in under three years. SGLang now serves 330+ organizations, supported 217 models on day 0, and runs on hundreds of thousands of GPUs. Sources include the LMSYS blog and SGLang Summit 2026 materials.
More from Infra
- JP Morgan private bank publishes note on Nvidia's Jensen Huang — BenBajarin · 2026-11-15
- Log screenshot shows aggressive crawler probing celesto.ai's wp-admin — aniketmaurya · 2026-11-15
- Back-of-envelope calc suggests frontier lab compute grids are plausible 10x scale — teortaxesTex · 2026-11-15
- Delip Rao asks: who is mirroring all of Hugging Face before it gets locked up? — deliprao · 2026-11-15
- Quantum software firm Q-CTRL raises $290M Series C, becomes Australia's first quantum unicorn — MJBiercuk · 2026-11-15
- One router at one site grounded every US flight: FAA outage was a config flood — emax · 2026-11-15