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] [AMD] diffusion: Pin VAE attention to math SDPA on all gfx11/gfx12
- [Feat] [AMD] sgl-kernel: Enable gfx11xx/gfx12xx (RDNA3/RDNA4) AOT builds
- [Test] sampling: cover softcap row argmax boundary and tie semantics
- [Fix] openai: Render DSV4.1 message tools with recipe
- [Feat] kv-cache: Mooncake v2 adds statistics for bandwidth and page count.
- [Test] mamba: cover hybrid radix checkpoints taken in mixed chunked-prefill forwards
- [Perf] [AMD] deepseek-v4.1: Run the HIP OPUS sparse prefill over MXFP8 rows with mla_sparse_prefill_v41
- [Perf] function_call: Drain Kimi-K2 arguments incrementally
- [Feat] [NPU] unlimited-ocr: Support Unlimited-OCR reference sliding windows on Ascend
- [Feat] observability: Expose HiSparse device and host cache metrics
- Docs
- Python not yet supported