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(glm47): require tool-call prefix for required selection
- [Feature] Upgrade Nsight Compute bundled by the CUDA devel base image
- [Scheduler] Skip the spec + DP prefill-first MLP sync while no rank has prefill work
- Fix DeepSeekV3 streaming: emit arguments when name and fence arrive in same delta
- [Fix] Stop at the earliest matching stop_str when several match
- [Docs] Align Ascend MemFabric install command with version table
- fix(spec): bind target embed/lm_head into the draft before KV pool profiling
- Development Roadmap (2026 Q4)
- [Model Loader] Configure tensor-copy concurrency for DeepSeek loaders
- [HiCache] Preserve ownership of pending file write reservations
- Docs
- Python not yet supported