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
- [Fix] pd: Isolate malformed NIXL decode staging requests
- [Bug] Profiling state conflicts return bare HTTP 500 responses
- [Feat] [NPU] attention: support FP8 KV with DSA context parallelism
- [Fix] Return HTTP 409 for profiling state conflicts
- [Feat] disagg: Let the PD prefill linker load without offloading
- [Fix] diffusion: Fix Cosmos 3 ignoring number of output per request for image and video requests
- [Perf] diffusion: Hy-Nomas fuse groupwise INT8 weight dequantization
- [Feat] [NPU] Use token-interleaved DCP KV for DSA and Kimi K3 MLA
- [Fix] HiCache: whole-page Mooncake IO with mixed-result preservation and staging safety
- [Fix] [XPU] graph: fake registration for optional MLA kernels
- Docs
- Python not yet supported