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
- [Spec] Move state capture outputs with the accepted path (EAGLE topk>1, ngram)
- Make prefill shared-read completion explicit and default-on
- [Bug] VisionTritonAttention drops MiMo-style window attention and sinks
- [Bug] Responses cancel endpoint accepts non-background responses
- [Fix] Reject cancellation of non-background Responses API requests
- [RFC] Quantization module layout: Config, Scheme, backend kernels
- [Bug] [DFlash2] Greedy output diverges from target-only Qwen3.8-27B when thinking is enabled
- [Bug] A single malformed multimodal request takes down the whole engine, and client retries spread it node-by-node across the cluster
- chore: bump docs install version to 0.5.19
- docs: fix outdated server flags in advanced-feature docs
- Docs
- Python not yet supported