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
- [Scheduler] Limit chunked prefill budget share with --chunked-prefill-ratio
- [Fix] move torch_native/flex_attention CUDA graph disable after platform fallback
- fix(attention): reject float32 model dtype at FlashInfer backend init
- [AMD] Select Quark MXFP4 dense GEMM via --fp4-gemm-backend aiter|triton
- Easymagpie/core fixes
- [MoE] Extend FlashInfer CUTLASS MoE autotune coverage to prefill chunks
- [MoE][gfx950] Pick the fp8_w8a8 default tile by routed rows per expert
- [AMD] Run Qwen3.5 GDN Triton decode on strided projection views
- [NPU] [DIFFUSION] Support Wan2.2-T2V-A14B W4A4F4 DAOS
- [Fix] Mooncake standalone: register MHA K/V at merged allocation granularity
- Docs
- Python not yet supported