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
- [quant] compressed-tensors: honor .linear-wrapped ignore entries on fused shards
- [MiniMax M3] Support EPD disaggregation
- [DSv4] Split draft SWA into committed sidecar KV and verify scratch
- [diffusion] feat: make the MiniMax-H3 ref2va reference-image short edge configurable
- [Fix] Monitor scheduler subprocesses owned by the DP controller
- [Bug] EAGLE speculative decoding (native Qwen3.5/3.8 MTP) hangs in tree_speculative_sampling_target_only on Ampere (2x A2, TP=2, 100% GPU util / low power)
- fix: skip singleton sampler token sync
- test: pin the int32 page-offset overflow in _page_split_kernel
- [MoonEP] Destroy communication buffers on runtime reset
- [P/D disagg] Decode-side radix cache for DeepSeek-V4 (DSA compressed KV)
- Docs
- Python not yet supported