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
- [CPU] Fix clang build: std::sqrt is not constexpr
- [Spec] Fuse aux hidden state capture and support DFLASH with breakable/full prefill CUDA graphs
- [Fix] multimodal: apply EXIF orientation when decoding images
- [Mamba] Fix wrong SSM state cached by the mamba radix cache under extra_buffer
- [PD][Spec] Skip spec-only aux slots from non-last PP ranks in send_aux
- [PD][MTP] Fix MHA V-pointer offset when target/draft KV layer counts differ
- [Bugfix][MoE] Make Marlin route alignment deterministic
- [NPU][FIX]: make Qwen3.6 radix-cache miss/hit lossless on Ascend A2
- [Model] Qwen2-MoE: overlap shared-expert compute with DeepEP communication on CUDA
- fix: Only top-two level keys are replaced, and inner nested dictionar…
- Docs
- Python not yet supported