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(lora): stop rebuilding the MoE align JIT module on every miss
- Fix deterministic chunked prefill starvation
- [Bug][NPU] ModelSlim W8A8 checkpoint of Qwen3.8-27B (GDN hybrid) loads with garbage weights: fused in_proj_qkvz scheme unresolved, vision encoder forced unquantized; NPU quant matmul crashes on 3-D vision input
- [Bug] Several ambiguous points in the installation documentation, please check
- fix: use colon separator for sglang metric names in Grafana dashboard
- [RFC] Resumable raw token stream for /v1/chat/completions
- [NPU] Make Ascend fused top-k/top-p sampling capture-safe and dtype-correct
- [Model] Add PLaMo3 model support
- [Bugfix] Initialize one_batch runtime configs in load_model
- fix(server_args): reject speculative decoding with torch_native attention backend
- Docs
- Python not yet supported