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: correct small text issues in server args docstrings and messages
- [Startup] Explain cold MXFP4 JIT builds when the optional FlashInfer cache is absent
- [Bug] `--default-chat-template-kwargs` with `reasoning_effort` silently overrides the per-request `reasoning_effort`
- [Speculative] Make DFlash device validation platform-extensible
- [Speculative] Add a device-neutral extend-cache location fallback
- fix(jit-kernel): exact fallback for fast_topk threshold-bin overflow (#36807)
- [Fix] Evict only the mamba slots a prefill batch will actually allocate
- [HiCache] Read host memory once per TP group before splitting the budget
- [Bug] GLM-5.3 DPC crashes
- [Bug] `fp8e4nv` not supported on A100 (SM80) when serving Qwen3.8-Flash-Next-FP8
- Docs
- Python not yet supported