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
- [UnifiedRadixCache] Bound SWA HiCache load-back to one sliding window
- Add native rawsystemone full-sequence scoring
- [DCP + L3 1/N] Support MLA DCP with HiCache file L3 across matching engines
- [Session] Fix admission and wait for all DP open acknowledgments
- [Session] Defer cleanup until in-flight requests finish
- [Bug] Diffusion Qwen-Image-2.1: TP=2 output corrupted with dense chroma speckle (single-GPU clean; vLLM-Omni TP=2 shows the identical corruption)
- fix: propagate weight checker failures across TP ranks
- [AMD][CI] Nightly guard for Qwen3.5-MoE text-only shared-expert fusion loading
- Fix processor BOS/EOS compatibility restoration
- [Sampling] Write /generate sampling masks from numpy rows
- Docs
- Python not yet supported