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
- [Feature] A general pattern for enabling piecewise CUDA graph on multi-modal models (Qwen3-VL / Qwen3.x)
- advance prefix_indices when skip_radix_cache_insert is True
- [NPU] Adapt DFlash for Ascend
- [gRPC] OpenAIServing.stream_chunks: skip SSE round-trip for non-HTTP transports (WIP)
- [Quantization] Support ModelOpt NVFP4 group-size configs
- [Bug Fix] Fix DeepSeek V2 forward when input_ids and input_embeds are missing on PP ranks
- [LoRA] Add sgl_flashinfer_trtllm MoE LoRA backend with two-stream overlap
- Improve startup plugin loading for external hardware platforms
- fix: add Rust cargo bin to runtime stage PATH
- LoRA Async Pipelining Per Requeust Load
- Docs
- Python not yet supported