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] Load compressed-tensors NVFP4 KV calibration with matched cache format
- [CPU] Add mamba_chunk_scan_combined and selective_state_update kernel on Xeon
- [Bug] DFlash + Qwen3.8-27B: speculative decoding accept rate drops to 0 during serving
- [Perf] [AMD] qwen3.5: Run each Qwen3.8 decode layer as two persistent FlyDSL launches on MI355X
- [Feat] [NPU] sampling: Add sampling-mask support for Ascend
- [Spec] Clear stale grammar masks before DFlash and DSpark verification
- [Roadmap][NPU][Multimodal] Model support and serving features (2026 Q4)
- [Feature] π0.5 completion-boundary batching for heterogeneous denoising budgets
- Fix duplicate global scaling in NVFP4 prefill workspaces
- [Feat] QSA: Add cache sharding for Qwen3.8 hybrid models
- Docs
- Python not yet supported