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
- EvalPort: portable interchange for BenchmarkResult
- [AMD] [Kimi-K3] Opt-in aiter fp8-Q asm MLA decode/verify
- [AMD] Run Quark MXFP4 MoE as W4A16 via triton_kernels on RDNA (gfx1151)
- Fix MiniCPMV.__getattr__ to preserve chaining on delegated methods
- [Model] Add Laya decision model (convaiinnovations/laya)
- fix(benchmark): avoid caching partial Mooncake downloads
- Fix Qwen3-Next weight loader double-fusing pre-fused gate_up_proj (fixes #13214)
- [AMD] Add experimental MiniMax-M3 FlyDSL attention
- [Bug] Accept pp_proxy_tensors in frozen-KV MTP and UNO spec-v2 workers
- [Bug] Reject dynamic LoRA updates when tokenizer-worker-num > 1
- Docs
- Python not yet supported