SGLang
Apache-2.0 inference framework for LLM, vision-language and diffusion models, optimized for agentic workloads and large-scale serving.
Tags
Description
Established open project (repo sgl-project/sglang; not archived; not presented as a 2026 launch). Root LICENSE is Apache-2.0; the PyPI license appendix names Copyright 2023-2024 SGLang Team. https://sglang.io 308s to https://www.sglang.io/ (HTTP 200); https://docs.sglang.ai 301s to https://docs.sglang.io/. This entry uses the final URLs. Docs https://docs.sglang.io/ and quickstart https://docs.sglang.io/docs/get-started/quickstart were HTTP 200 on 2026-11-14. https://cookbook.sglang.io/ 302s to https://docs.sglang.io/cookbook/ and is not used as a URL or source. PyPI sglang 0.5.23 (not yanked; wheels uploaded 2026-10-30T01:34:33Z = 2026-10-30 02:34:33 Europe/Stockholm and following). GitHub latest release v0.5.23 was published 2026-11-02T22:14:10Z = 2026-11-02 23:14:10 Europe/Stockholm (not draft/prerelease). README documented install in an activated environment: uv pip install --prerelease=allow sglang. That --prerelease=allow flag is as written in the README even though 0.5.23 is a non-yanked PyPI release; the command was not executed. README also describes a Docker path from lmsysorg/sglang:dev and contact sglang@lmsys.org for enterprise adoption. Model weights and any GPU or host resources are licensed and billed separately.
Agent access
Related
- vLLM: Apache-2.0 high-throughput LLM inference and serving engine with an OpenAI-compatible API, installable via uv or pip.
- llama.cpp: MIT C/C++ library and server for local LLM inference, widely used as a local OpenAI-compatible backend for coding agents.
- Ollama: MIT app and API for running open models locally, plus a paid Ollama Cloud with usage credits and per-token prices.
- LocalAI: MIT self-hosted OpenAI-compatible engine that runs LLM, vision, voice and image backends locally, with Docker images as the documented install.
- MLX LM: MIT Python package for text generation and fine-tuning of large language models on Apple silicon with MLX.
Guides & comparisons
- Run an agent on a local model with an OpenAI-compatible endpoint (Guide): Start Ollama, vLLM, SGLang or LocalAI bound to 127.0.0.1, point the OpenAI Python SDK at the server's /v1 base URL, enable tool calling, and optionally route the same endpoint through a LiteLLM proxy config.
- Fully local, private agent (Stack): A local OpenAI-compatible model server, an optional LiteLLM router, Pi as the harness, Basic Memory for on-disk Markdown memory, and Anthropic's sandbox runtime — all on your machine, with no hosted model API and no cloud sandbox.
Sources
- https://www.sglang.io/
- https://github.com/sgl-project/sglang
- https://github.com/sgl-project/sglang/blob/main/LICENSE
- https://github.com/sgl-project/sglang/blob/main/README.md
- https://github.com/sgl-project/sglang/releases/tag/v0.5.23
- https://docs.sglang.io/
- https://docs.sglang.io/docs/get-started/quickstart
- https://pypi.org/project/sglang/
- https://pypi.org/project/sglang/0.5.23/
Machine-readable
- JSON
- Markdown
- Source file on GitHub (edit via pull request)