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] SenseNova-U1: let think_mode image tokens attend to the reasoning
- [Feat] [NVIDIA] fp8: Add an H20-3e block-32 FP8 GEMM for DeepSeek-V4.1-Flash decode
- [Feat] hicache: Support file storage with MLA host dedup
- [Fix] benchmark/mmmu: Forward extra request body through the SDK
- [Feat] hicache: Support DeepSeek V4 multi-pool host dedup
- [Fix] Minimax-M3: Deterministic tie-breaking in the CUDA radix decode top-k
- [Feat] hicache: Deduplicate MLA layers in Kimi hybrid stacks
- [Feat] hicache: Honor dist timeout during MLA dedup group setup
- [Feat] hicache: Extend shared host reads to V4 and hybrid MLA pools
- [Feat] diffusion/minimax-h3: Add an online FP8 video VAE decoder
- Docs
- Python not yet supported