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
- [Kernel] Avoid batch-size specialization in masked KV writes
- Fix fa4 crash with paged KV diff-headdim split-KV; guard MiMo SWA 0-token
- Add optional realtime TAEHV decode
- [MLX] Fix hybrid-SSM + radix cache crashes in the auxiliary-state component
- [PD] Support MiniMax-M3 sparse KV CPU copy
- [Fix] Report DCP logical token capacity in startup log and metrics constant
- :bug: [NPU] Fix Qwen3.5 dense serving on Ascend 950
- Fuse GDN decode projection unpack with Conv1D
- GLM5.2 DSA Indexer Fallback
- [AMD] Add AITER HIP backend for packed GDN decode on gfx950
- Docs
- Python not yet supported