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
- [Feat] bailing: Add option to disable router gate matvec
- [Fix] diffusion: Fall back to the cached snapshot when the Hub API lookup of weights fails
- [Doc] [NPU] cookbook: Add Ascend A3 deployment scenarios for DeepSeek-v3.2
- [Perf] SANA: dispatch strided fused LayerNorm
- [Fix] pp: Materialize nested output metadata before serialization
- [Feature] moe: FlashInfer cuTile NVFP4 MoE as the default runner on SM12x
- [PD] Fix PD + Mooncake linker on a shared transfer engine by sharing its memory registrations
- [Feat] scheduler: Expose single-iteration API
- [Fix] hicache: reject short Mooncake zero-copy reads
- [Feat] [Ascend] Integrate A5 MXFP experts with MegaMoE
- Docs
- Python not yet supported