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
- Profiling Enhancements [3/3]: Shape registeration for kernels in SGLang
- [NIXL] Doca memos support
- [Observability] Break down prefill #cached-token into device / host / storage tiers
- benchmark: add MI300X speed benchmarks for Qwen3.5-4B cookbook
- [DSV4] Enable overlap scheduling for online C128 MTP
- [Scheduler] Reuse per-step cuda events uniformly (WAR read_done + copy_done)
- [Responses] Add output logprobs support for harmony and streaming
- [NPU] Fix ModelSlim NEXTN startup: unquantized MoE fallback and UB-aware mamba scatter
- [Fix] Fix hybrid-SSM DP attention failures at low concurrency
- Fix overflow in legacy and v1 DSA top-k selectors
- Docs
- Python not yet supported