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
- [Test][NPU] overlap-schedule validation: GLM5/Kimi/DSV4 case reconfig + TEMP nightly scoping
- [Perf] [NPU] chunked-prefill: KV cache pruning for extra-long context
- [Fix] hicache: validate decode L3 hits and preserve transferred Mamba state
- [Doc] kernels: correct the documented output dtype parameter name
- [Doc] srt: restore docstring parameter names the signatures no longer take
- [Fix] openai: Honor chat-template stop strings for pre-tokenized input_ids
- [Perf] AMD Follower of allreduce auto-tuner from PR 34838 by qunatised QRs
- [Feat] pp: Relay routed experts through pipeline-parallel stages
- [Feat] model: Add Kimi-K3.1 support
- [Feat] [NPU] kimi-k3.1: Add model support and route DSA attention path on Ascend
- Docs
- Python not yet supported