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
- [Feat] logging: add SGLANG_LOG_REQUEST_SKIP_FIELDS to elide large request fields
- [Feat] diffusion: abort queued rollout requests
- [Fix] [NVIDIA] kda: Invert diagonal A_kk blocks in fp32 to fix prefill NaN
- [Feat] diffusion: support dynamic multi-LoRA in ComfyUI
- [Fix] openai: Align DeepSeek DSML streaming unknown-tool filtering and tag prefix guards (#44941)
- [Feat] ngram: let each request choose which external corpora to draft from
- [Refactor] platforms: in-tree platforms declare their BaseFusedOp dispatch key
- [NPU] Please keep publishing Docker images for Ascend 910B
- [Refactor] platforms: in-tree platforms provide their graph runner and piecewise backend
- [Refactor] quantization: ask the platform for its quantization config on every device
- Docs
- Python not yet supported