AITOPAI热报
ProductsSelected

SGLang 三年进化：从 RadixAttention 到支撑 330+ 组织的推理引擎

AI Will@FinanceYF5
Post
View the post
Why we picked it

想自己搭推理服务的开发者可以看看这条时间线。SGLang 从 RadixAttention 缓存技巧做起，三年内做到 217 个模型 day-0 支持，330 多个组织在用，和 vLLM、TensorRT-LLM 属于同一赛道。选引擎前了解它的演进路径会有帮助。

LMSYS 于 2026 年 11 月发布 SGLang 演进时间线，梳理了三年内的 20 个里程碑。项目起点是 2024 年 1 月提出的 RadixAttention 缓存技术，到 2026 年 11 月已发布 v1.0 版本并成立 SGLang Consortium。

Image · AI Will
Body

目前 SGLang 被 330+ 个组织采用，支持 217 个模型的 day-0 接入，运行在数十万张 GPU 上。技术路径上，它从最初的前缀缓存技巧逐步扩展为完整推理引擎，覆盖模型服务、调度与并行执行。

素材来源包括 LMSYS 官方博客、@sgl_project 于 2026 年 11 月 12 日的发布、SGLang Summit 2026 的时间线墙与主题演讲。对一个从研究项目成长为大规模生产级推理引擎的开源项目而言，这条时间线展示了社区共建模式的落地过程。

Original · AI Will

The Evolution of SGLang, 2024 → ∞: from a caching trick to the inference engine behind 330+ organizations | 2026 Under three years, twenty milestones: from RadixAttention (Jan 2024) to SGLang v1.0 and the SGLang Consortium (Nov 2026). Today SGLang is used by 330+ organizations, supported 217 models on day 0, and runs on hundreds of thousands of GPUs. Music only. Every frame and every note generated in code. Sources: LMSYS blog and @sgl_project (Nov 12, 2026), @lmsysorg (Nov 11, 2026), @radixark (Nov 12, 2026), SGLang Summit 2026 timeline wall and keynote slides. AI-generated content (视频由 AI 生成).

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Products" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include SGLang, LMSYS, RadixAttention, inference engines.

Source attribution: curated from AI Will (@FinanceYF5).

#SGLang#LMSYS#RadixAttention#inference engines#open-source models

Keep reading

Recent stories with the same tag or section
PapersBuildersMulti-source04:00
Chinese contentCMU 论文：Opus 5.5 + Claude Code 搭建的系统在 Singularity-Bench 上超越 SGLang

CMU 这次不考答题了，直接让智能体搭整套生产系统跟 SGLang、NVIDIA Dynamo 硬碰硬。Opus 5.5 做出的推理引擎每 GPU 小时赚 11.52 美元，比 SGLang 的 8.18 还高。做智能体评测和基础设施的工程师都该看看评分方式。

#Opus 5.5#Claude Code#SGLang#Singularity-Bench
Event
rohanpaul_ai@rohanpaul_aiSource
IndustryBuildersSelected02:26
Databricks Takes 2nd in Meta Muse Spark Kernel Challenge

Autonomous agent loops are now producing production-grade GPU kernels that beat mainstream inference-framework baselines, which lowers the cost floor for serving AI workloads. Engineering teams running high-volume inference can treat LLM-driven kernel optimization as a practical pipeline rather than a research demo. The finding that Opus 5.5 outperforms other models on kernel optimization also gives practitioners a concrete model choice for similar code-optimization agents.

#Databricks#Claude Opus 5.5#SGLang#GPU kernel optimization
Yuchen Jin@Yuchenj_UWSource
IndustryBuildersMulti-source00:46
Chinese contentNVIDIA 与 SGLang 社区合作：跨五代硬件、60+ 模型零日支持

SGLang 和 vLLM 是目前自部署大模型最常用的两个推理框架，NVIDIA 这次晒出五代硬件、60 多个模型的零日适配数据，说明对 SGLang 的工程投入是真金白银。自己搭推理服务的开发者可以关注后续 Summit 的技术分享。

#SGLang#NVIDIA#vLLM#open-source models
Event
Official accountNVIDIA AI@NVIDIAAISource
ProductsBuildersOfficial06:27
Chinese contentSGLang-Neuron 上线 AWS Deep Learning Container，支持在 Trainium 上部署大模型

SGLang 用户现在可以直接把推理和 RL rollout 跑到 AWS Trainium 上，接口不变、原生 PyTorch、AWS 官方容器开箱即用。想在 GPU 之外降推理成本或统一训练推理硬件的团队可以试试这条路。

#SGLang#AWS#Trainium#inference deployment
Event
Primary sourceAWS Machine Learning Blog@Alex NailsSource
ProductsBuildersMulti-sourceSelected06:06
SGLang-Neuron brings SGLang serving to AWS Trainium

Serving stacks have mostly been GPU-bound, so a Trainium path for SGLang reduces the engineering cost of running inference on AWS's own silicon. Teams standardising on SGLang for RL rollout generation can now target Trainium instances directly. This matters for anyone evaluating Trainium against NVIDIA GPUs for cost or supply reasons, since the same APIs carry over.

#SGLang#AWS#Trainium#inference deployment
Event
Official accountAmazon Science@AmazonScienceSource
IndustryBuildersMulti-sourceSelected02:57
SGLang Consortium formed with NVIDIA, AMD, Google and Meta

Hardware vendors co-funding a single open-source inference engine reduces the risk that serving optimizations remain locked to one chip ecosystem, which matters for teams planning multi-vendor GPU fleets. Engineering leads running large-scale inference can track the consortium's roadmap to anticipate support for new models and distributed serving features. Organizations in Southeast Asia building inference capacity on heterogeneous hardware will benefit from upstream optimizations landing in SGLang rather than vendor-specific forks.

#SGLang#NVIDIA#open-source models#inference frameworks
Event
Official accountLMSYS Org (SGLang)@lmsysorgSource