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
- chore: bump docs install version to 0.5.17
- scheduler: bound consecutive prefills to protect decode
- feat(moe): add opt-in FlashInfer AlphaMoE W8A8 and NVFP4 backends
- [Feature] Support config.json from transformers > 4.57 for DeepSeek V4
- flashinfer: pass logits_soft_cap (and ragged window_left) at plan time; gemma-2 / grok ran uncapped
- [Bug] A cancelled grammar-constrained overlap request can still emit visible text before abort
- [Doc]The invalid links in the document
- [router] Add bucket range config parsing, validation, and selection h…
- [Fix] Give multimodal draft-extend one mRoPE position per extended token
- [AMD] Port CP V2 to the DeepSeek-V4 HIP backend
- Docs
- Python not yet supported