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
- [Perf] EAGLE rejection sampling on each row's top-k support
- Fix/compressed tensors mxfp4 marlin
- Fix/compressed tensors mxfp4 moe
- [Fix] Feed the relayed last token to penalizers under the overlap scheduler
- Honor OOT attention defaults for Gemma4
- [Perf] Use the fused low-ratio compressor on Hopper
- [Perf] Skip unused K/V work in Gemma3n shared attention
- [Bug][ROCm] Qwen3.8-Flash-Next (qwen4_exp) crashes with HSAIL hardware exception 0x1016 on the first 16k prefill chunk, gfx950 TP1
- [quantization] Add tuned W8A8 block-FP8 GEMM configs for NVIDIA H20 (block_shape 32x32)
- docs: keep multi-node example with its introduction
- Docs
- Python not yet supported