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
- docker: bind-mount builder artifacts instead of COPY + rm (374 MB of committed staging layers)
- [GGUF] Qwen3.5 GDN GQA head expansion + merged-layer shard loading
- [Perf][DSA] Pass transform_index strides at runtime instead of constexpr
- [Fix][Spec] Bound draft-extend replay metadata to graph buckets
- [Bugfix] Stop video_decoder truncating and orphaning its temp file
- [Docs][AMD] Refresh Qwen3.8-Flash-Next ROCm recipes
- Rainj me/rust api server tower layer
- [Bug] --enable-symm-mem: all TP schedulers deadlock silently in ncclCommWindowRegister (blocking UDS recvmsg, no timeout) when the symmetric-memory pool grows at runtime
- [zaya] ZAYA1 (CCA attention + MoE) serving path (3/3)
- fix(npu): skip FLA autotune during the K3 first request
- Docs
- Python not yet supported