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
- [Bug] GLM-5.3-Flash: engine state corrupted after a 1M-token request on v0.5.23 (MTP accept rate 0.00, all subsequent requests degenerate); v0.5.22 clean
- [Fix] gemma4: Fix infinite loop in tool-call array parser
- [Bug] [AMD] decode attention: MLA-shaped Triton decode returns NaN / wrong rows on ragged batches (gfx1151)
- [Perf] [AMD] decode-attention: Skip the CDNA launch hints on RDNA3/3.5/4 for GQA decode
- [Fix] function_call: support Python reserved keywords as parameter names in PythonicDetector
- [Fix] health_check: Require completed inference for /health_generate, not any scheduler message
- [Feat] nemotron_labs_diffusion: Add self-speculative decoding
- [Perf] Hunyuan3D: Speed up QEM mesh decimation (Lossless)
- [Fix] [NPU] disaggregation: Preserve MLA KV layer mapping across PP stages
- [Doc] sampling: Clarify CUDA greedy ties and LM-head output precision
- Docs
- Python not yet supported