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
- docs(cookbook): move rolling :dev pages to :latest where support shipped in v0.5.15
- Fix abort request counter accounting for abort_all
- fix(function-call): stream DeepSeek-V3 tool-call args when whole call in one chunk
- fix(router): count shared prompt tokens once for n>1 in gRPC chat streaming
- [Fix] Hoist $defs for named tool_choice JSON schema constraint
- [Bug] Hybrid-GDN models: mamba conv_states dtype ignores --dtype; float16 serving crashes at first prefill with index_put dtype mismatch
- ci: upload runner-utilization report as a run artifact
- feat: add per-position speculative acceptance metrics
- How should I configure the service to make the model return tool calls that comply with the OpenAI standard instead of XML format?
- [Bug] fd exhaustion on the prefill side
- Docs
- Python not yet supported