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
- [Perf] logprobs: Cut latency and memory with compact TP exchange
- [Perf] ngram: Freeze external-corpus SAMs into a compact layout
- [AMD] Add MiniMax-M3 dense MXFP8 FlyDSL GEMM
- [Perf] multi-tokenizer: Group MultiTokenizerRouter outputs by tokenizer worker
- [Perf] request-logger: Skip logged fields before watermark redaction
- [Bug] WeightChecker reset_tensors fills MRotaryEmbedding.axis_map with random values
- [Fix] tokenizer: Offload regular tokenization off the event loop
- [Bug] update_weights_from_disk reports success when Qwen3.5-VL vision weights have unknown names (no warning for .bias)
- [Fix] function-call: preserve quoted brackets in Pythonic streaming
- [Fix] [NPU] glm-ocr: Fix M-RoPE section for GLM-OCR on Ascend and add A3 validation
- Docs
- Python not yet supported