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] gemma4: Support --language-model-only
- [Feat] hicache: Add a Mamba/KDA L3 snapshot interval
- [CI] sail: add a release workflow for the sglang-kernel wheel
- [RFC] Kimi K3 decode: FlashInfer reuse and integration scope
- [Doc] cookbook: Point the omni FishAudio card at the route Mintlify serves
- [Feat] spec: Add EXTERNAL speculative decoding algorithm
- [Perf] [NVIDIA] diffusion/joy: Reuse FP32 residual gate for large image updates on SM103
- [Perf] model_loader: Stage weight copies to the GPU through pinned memory
- [Perf] dcp: pack each peer's send slots from one program
- [Bug] Qwen4-Exp PLE: BF16 n-gram shards are cast to FP8 with a unit scale when the checkpoint's config sets ple_embedding_dtype (primitive-ai/Qwen3.8-Flash-Next-NVFP4)
- Docs
- Python not yet supported