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] Preserve PD bootstrap fields in Anthropic requests
- [Intel XPU] Add a sanity cap to free xpu memory
- [AMD] Recalibrate the Quark MXFP4 MoE GSM8K threshold 0.92 -> 0.89
- fix(overrides): default moe_runner_backend to flashinfer_trtllm_routed for minimax-m3 nvfp4 on b200
- fix: bound FlashInfer DeepSeek routing shapes
- [Bug] Session continuation crashes the engine on M-RoPE models: unguarded partial slice of the mm-input M-RoPE position table
- Fix M-RoPE positions when an extend reaches past the precomputed mm table
- fix: handle multimodal Kimi-K3 Responses prompts
- [DSV4] unified_kv: isolate C4 compress state per request
- [Feature] Add a tiered AdaLN cache manager for MiniMax-H3
- Docs
- Python not yet supported