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
- [CI] debug_utils: Set test_dumper nightly est_time from measured runs
- [Perf] diffusion/attention: Fuse VSA tile packing and layout conversion
- [Perf] speculative: Fewer launches in a top-1 EAGLE decode step
- Anmolgupt/nvfp4 silu quant fusion
- [Fix] profiling: Require bounded per-stage profiling
- [Fix] server_args: Reject non-positive prefill request limits
- [Fix] Spec: Fall back to eager when the ragged-verify tier exceeds the captured grid
- [Fix] grpc: Set TCP_NODELAY on accepted runtime.v1 gRPC connections
- [Perf] [AMD] GLM5.3: Allow AITER MXFP4 scatter intermediate only for GLM-5.3-Flash on ROCm
- [Fix] model_loader: Neutral UE8M0 block scales for dummy weights
- Docs
- Python not yet supported