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] attention: Isolate GDN MIS item state from delimiters
- [Fix] attention: Synchronize Qwen3.5 text MIS MRoPE positions
- [Fix] [AMD] fp8: Respect token padding in the native static quant fallback
- [Bug] HiCache L3 (Mooncake) backup fails for GDN models with PLE slot side states (L3 hit rate stuck at 0)
- [Fix] HiCache: whole-page Mooncake IO for mamba pools with PLE slot side states (L3 hit rate stuck at 0)
- [Perf] [NVIDIA] attention: accelerate DOTS3 MLA/SWA decode with FlashInfer on Hopper
- [Perf] qwen3.5: Support prefill MBO (stacked on #43655)
- [Fix] disaggregation: align DSA tail seq_len with decode on last chunk
- [Perf] deepgemm: Fuse SwiGLU activation and FP8 quantization
- [Fix] [AMD] model-loader: register full allocations for remote weight loading
- Docs
- Python not yet supported