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] Use int64 token offsets in the fused QK RMSNorm + RoPE + gate kernel
- Add opt-in GLM MXFP4 MoE operators for four-token verification at C1/C2
- [PD] Shard Mooncake transfers by bootstrap room
- [DSA] Read the MQA logits budget once per forward batch in the k-pool indexer
- Accept pp_proxy_tensors in the frozen-KV MTP and UNO spec workers
- [8/N] [HiSparse][PD] Bound a PD HiSparse decode's requests by its logical KV pool
- [HiCache][AMD] Move page_first_direct pages with the gather kernel on ROCm
- RFC: Capability-based device gating for CUDA-compatible OOT platforms
- [Feature] Integrate Cake kernels via FlashInfer: model-by-model tracker
- [Fix] Normalize nested thinker text config dictionaries
- Docs
- Python not yet supported