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] scheduler: Skip a chunked prefill when the KV pool holds no free page
- [Fix] diffusion: Return latent_step_indices and keep every timestep when rollout filters trajectory steps
- [Feat] clef: Honor an explicit prefill CUDA graph backend
- [RFC]: add eagle4 spec decode
- [Bug] Kimi K3 tool-call parser (kimi_k3) does not stream tool-call arguments: whole call is buffered until <|close|>call, leaving the stream silent for minutes on long tool calls
- [CI] [NVIDIA] gb300: Remove the 4-gpu-gb300 CI suite
- [Fix] [AMD] dflash: Sample-verify temperature > 0 drafts on ROCm instead of falling back to greedy
- [Feat] [AMD] DeepSeek-v4.1: Support HiSparse V2 with HiCache-backed compressed KV
- [Feat] GLM-5.2: Support NVFP4 KV cache with DSA layer split
- models: deepseek 4.1 flash perf improvements
- Docs
- Python not yet supported