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] [NPU] cache: enable compact DSA indexers for PD and HiCache
- [Fix] [NPU] deepep: support Normal BF16 with quantized LowLatency
- [Perf] io_struct: Validate an input_embeds body against one union member
- [Feat] disaggregation: Add optional KVServe compression to Mooncake P/D
- [Perf] dsa: Add an opt-in FlashInfer Prims-TS sparse MLA backend
- [Perf] [NPU] DeepSeek-V4: remove blocking H2D copies and fuse DSPARK verify compressor-position metadata into one triton kernel
- [Fix] dspark: Count scheduler ranks in SPS profiler
- [Feat] scheduler: Add waitable request retirement receipts
- [Fix] rl: One tensor weight payload per engine rank under pipeline parallelism
- [Feat] model-loader: ModelExpress weight release, resume and reload hooks
- Docs
- Python not yet supported