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
- perf(openai): keep prompt_ids for text-only requests on multimodal checkpoints
- fix: honor MiniMax thinking controls
- [Feature] [RFC] SharedEP: a zerocopy execution model for MoE Expert Parallelism
- [Perf] HiCache storage prefetch finalization delay can cause long TTFT under load
- [Feature] Add KV event gap replay for experimental sgl-router
- [Feature] PP Support PD + DSpark
- [Router] Wire decode-affinity outcome metric
- [Feature] Add optional admin auth for sgl-router /flush_cache
- [Router] Add optional admin auth for /flush_cache
- TX PFC / NIC pause frames during Mooncake L3 cache prefetch — NIC→host-memory intra-node bottleneck
- Docs
- Python not yet supported