sglang
https://github.com/sgl-project/sglang
Python
SGLang is a high-performance serving framework for large language models and multimodal models.
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported1 Subscribers
Add a CodeTriage badge to sglang
Help out
- Issues
- [Fix] Harden DeepSeek-V4 tool-call streaming
- fix(moe): don't crash on a2a dispatch outputs without topk_output in mxfp4 path
- fix(moe): size mori/pplx dispatch-token checks for the DP-attention gather
- fix(openai): accept a single tool-call object under tool_choice=required
- [Scheduler] Concurrent chunked prefill via --long-prefill-token-threshold
- [Scheduler] Remove GPU rendezvous from symmetric DP metadata sync
- [Bug] Guided decoding resumes one token too early for channel-framed reasoning formats (Muse Glimmer / ATEM)
- Support pre-expanded Qwen3-VL video inputs
- fix: reasoning_tokens counted when no thinking block is opened
- [AMD][DI][Mooncake] Cap free GPU memory to guard against KFD PeerDirect accounting corruption
- Docs
- Python not yet supported