SGLang
SGLang is a runtime maintained by LMSYS (hosting organization)27 and RadixArk (company described as a maintainer in a post on the LMSYS blog)910, version 0.5.20, released Sep 18, 2026. Its source code is public123 under the Apache License 2.03.
SGLang is an open-source serving framework for large language models and multimodal models, designed for low-latency, high-throughput inference on anything from a single GPU to large distributed clusters. Its runtime features include RadixAttention prefix caching, prefill-decode disaggregation, speculative decoding, continuous batching, several forms of parallelism, structured outputs, and quantization.2
- Repository: GitHub repository (external site: github.com)
- Documentation: Installation guide (external site: docs.sglang.io)
- Release notes: Release v0.5.20 (external site: github.com)
- License: License (Apache 2.0) (external site: raw.githubusercontent.com)
Availability and license
Component reuse rights
- weights
- Unknown — no complete fact-level rights review
- code
- Unknown — no complete fact-level rights review
- data
- Unknown — no complete fact-level rights review
- documentation
- Unknown — no complete fact-level rights review
No complete system-rights review is recorded for this release.
Public materials checklist
| Item | Status | Notes and evidence |
|---|---|---|
| Source codeIs the source code publicly readable? | Public | Published on GitHub under the Apache License 2.0.123 |
| DocumentationIs user documentation published? | Public | The documentation includes installation, quick-start, backend and frontend tutorials, a per-model deployment cookbook, and a contribution guide.24 |
| InstallationAre installation instructions or packages publicly available? | Public | Documented methods include pip or uv, installing from source, Docker images, Kubernetes, and cloud deployment options.4 |
| Supported platformsAre supported operating systems or hardware documented? | Public | The install guide mainly covers NVIDIA GPUs and has separate pages for AMD GPUs, Apple Metal, Intel Xeon CPUs, Intel XPU, Google TPU, NVIDIA Jetson and DGX Spark, and Ascend NPUs.42 |
| Release statusAre versioned releases published? | Public | Versioned releases are published on GitHub and PyPI; v0.5.20 was released on 2026-09-18. Nightly wheels are also available.564 |
What it is useful for
Serving language, embedding, reward, and diffusion models behind an OpenAI-compatible API, and acting as the rollout backend in reinforcement-learning post-training frameworks.2
Run and use notes
- The installation guide requires Python 3.10 or higher. It states that SGLang requires CUDA 13, that the CUDA 12 wheels and images are retired, and that 0.5.19 was the last release with CUDA 12 builds.4
Organization context
In the news
SGLang 0.5.22 adds Apple Silicon support with the PyTorch backend and an InstantTensor loader
The SGLang project published v0.5.22 on GitHub on October 19, 2026. The release notes say SGLang now runs on Apple Silicon Macs with the standard PyTorch backend, alongside MLX, that a new InstantTensor loader speeds up model loading, and that HiSparse now supports EAGLE and MTP for long-context serving with the KV cache in CPU memory. Newly supported models listed include Clef / Clef Flash, PPLX-Decider v1 / v1.1, EmbeddingGemma 2, Kandinsky 6 and SANA-Video 2.0. The notes say SGLang now requires Python 3.11 or newer, and describe a fix for hybrid linear models that could resume from wrong mamba states when speculative decoding and structured output were used together. USASI has not tested these changes.
SGLang 0.5.21 adds a Decisions API and runtime prefill-decode role switching
The SGLang project published v0.5.21 on GitHub on October 2, 2026. The release notes highlight prefill-decode instances that can switch between prefill and decode without a restart, a Rust core for the prefix cache by default, a new Decisions API (
/v1/decisions) that lets an LLM or VLM act as a low-latency classifier and scorer, and a Score API (/v1/score) that scores all candidates in one request. Under breaking changes, request-suppliedchat_templatevalues are rejected unless the server starts with--trust-request-chat-template, and deprecated endpoints, environment variables, and aliases, along with legacy radix cache implementations, were removed. The release also adds support for models including DeepSeek-V4.1 Flash and MiMo-V2.6. USASI has not tested these changes.
U.S. eligibility
Eligible · basis: U.S.-governed project
The SGLang README states that the project is currently hosted under the non-profit open-source organization LMSYS. LMSYS describes itself as a 501(c)(3) nonprofit incorporated in September 2024. The IRS extract for California lists LMSYS Corp, in Sacramento, as a 501(c)(3) corporation with a September 2024 ruling date. A July 2026 post on the LMSYS blog, credited to RadixArk and Google, also describes the company RadixArk as a maintainer; this assessment rests on LMSYS hosting, and RadixArk's location was not verified. SGLang joined the PyTorch Ecosystem in 2025, but the pages read do not describe it as a PyTorch Foundation-hosted project.278911
Sources
This listing is not an endorsement, a safety assessment, or a federal approval.
Share or cite this page
- X (opens X)
- Bluesky (opens Bluesky)
- LinkedIn (opens LinkedIn)
- Reddit (opens Reddit)
- Hacker News (opens Hacker News)
- Email (opens your email app)
- RSS updates
Cite this page
APA style
USASI. (2026, September 29). SGLang. United States of America Superintelligence. https://unitedstatesofamericasuperintelligence.com/open/sglang/
BibTeX
@misc{usasi_sglang,
author = {{USASI}},
title = {SGLang},
year = {2026},
month = sep,
howpublished = {\url{https://unitedstatesofamericasuperintelligence.com/open/sglang/}},
note = {United States of America Superintelligence. Last updated 2026-09-29}
}The date is when this page was last updated. Add the date you read it if your style needs one.
For maintainers: a “Listed in USASI” badge
Links to this page. It says the entry is listed, not endorsed or rated. About the badge
Markdown (for a README)
[](https://unitedstatesofamericasuperintelligence.com/open/sglang/)
HTML
<a href="https://unitedstatesofamericasuperintelligence.com/open/sglang/"><img src="https://unitedstatesofamericasuperintelligence.com/badges/listed.svg" alt="Listed in the USASI catalog" height="20"></a>