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
- [Feature] Add clef serving support
- [PDMux] Support GLM-5.3-Flash with layerwise prefill and overlap
- [PDMux] Reserve Mamba and HiCache state across prefill
- [PDMux] Align GLM DP and EP execution across PDMux streams
- [PDMux] Add GLM single-layer MTP support
- [cake_kernels] opt-in P1/P2 model routes: DeepSeek-V3 grouped routing, DeepSeek-V4 sparse MLA decode, Mamba2 SSD/SSU, SP all-gather matmul, MiniMax-H3 diffusion stages (SGLANG_CAKE_ROUTES)
- fix: include tool calls section in DeepSeekV31 structure info
- fix: handle disabled expert distribution recorder gracefully
- [diffusion] Reset Spectrum state per CFG branch
- [Spec] Reject --speculative-token-map when the target lm_head is vocab-sharded
- Docs
- Python not yet supported