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 InternlmDetector drops streamed tool calls after the first
- Fix DeepSeek-V4.1 DSpark decoder-tail run_idle row count mismatch
- fix(function_call): support streaming multiple tool calls in InternlmDetector
- [Feature] [Model Support] Native SGLang Support for MemT-1.0 (Nano and Flash)
- [Feat] remote-instance: Publish one rank-info record per runner rank
- [Feat] weight-sync: Add update_weights_from_colocated_param_bytes
- [Fix] cuda-graph: Keep attention metadata and SP workspaces alive across graph variants of a batch size
- [Test] cuda-graph: Add a standalone CUDA-graph lifetime validation tool and its skill
- [Feat] metrics: expose effective request capacity
- [Perf] deepseek-v4.1: Feed encoder-only decode tails through DSpark verify steps
- Docs
- Python not yet supported