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] speculative: Exact NEXTN penalties and add opt-in CORAL H100 overlay
- [Fix] diffusion: Import fp8_fa_sm120 under a meta-device model build
- [Feat] [NVIDIA] diffusion: Enable fp8_fa_sm120 attention on SM89
- [Fix] function_call: keep a stray closing function tag inert in qwen3_coder streaming
- [Refactor] kda: Import recurrent_kda from the canonical flashinfer namespace
- [gRPC] Add lifecycle-aware standard health checks
- [Feat] kernels: Beam model support
- [Perf] function_call: Optimize Gemma4, Hunyuan, GigaChat3, and GigaChat35 streaming tool detectors to O(N)
- [Perf] [NVIDIA] fp8: Use FlashInfer b12x for block-fp8-as-MXFP8 linears on SM12x
- [Perf] [NVIDIA] mhc: Use BLOCK_K=32 for the hc_mix kernel on SM120
- Docs
- Python not yet supported