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
- [Feat] [AMD] minimax: gfx950 MoE, router GEMV, block-fp8 GEMM and fp8 producer kernels
- [Feat] [AMD] distributed: JIT quick all-reduce for TP4 on gfx950
- [Bug] qwen3.8-27b says "let me look at the file" but emits no tool call, then concludes without evidence (thinking-only / tool-call dropped)
- [Feat] model-loader: Support per-model Hub source prefixes
- [Feat] distributed: Use pre-created device groups for an engine inside a larger world
- [Perf] openai: Skip the prompt encode/decode round trip for multimodal chat (opt-in)
- [Perf] tokenizer_manager: Optionally wrap SHM features off the event loop
- [Fix] gdn: Honour the fp32 radix track snapshot in Triton extend
- [Perf] GLM-5.3: Restore TP decode fusion and graph sizing with CP prefill
- [Perf] srt: Replace torch.nonzero() with torch.where() for index extraction (#9889)
- Docs
- Python not yet supported