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
- [Feat] dsv4.1: [4/5] Support CP>1 - add P/D replay handoff for DSpark
- [Feat] dsv4.1: [5/5] Support CP>1 - add Mooncake L3 recovery
- [Fix] hicache: Move NIXL pages that are not whole O_DIRECT blocks
- [Fix] dsv4: pass the mHC sinkhorn residual argument explicitly for torch.compile
- [Feat] mem_cache: Add programmable KV cache control and composable policies
- [sglang-miles] [Diffusion] Return each sample's decoded audio in the rollout response
- [sglang-miles] [Diffusion] Let a rollout request skip the VAE decode
- [Fix] PD/PP: Fix consensus checker failure while using PP prefill
- [Fix] anthropic: Keep K3 token counts consistent with Messages usage
- [MUSA][help] models support matrix on the MUSA platform
- Docs
- Python not yet supported