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
- [MoE] Return softmax probabilities from raw-logits topk when renormalize=False
- [Fix] shared expert under-weighted when rsf is pre-folded in topk weights
- [Model Gateway] Publish gRPC breaker outcomes at selected body terminal
- [Fix] Keep normal text preceding the tool-call opener in streaming parsers
- [Fix] Extend RoPE cos/sin cache with the subclass inv_freq and row scaling
- test(npu): add glm5_1 w4a8 32p in16k perf case and nightly NPU workflow updates
- fix: fix: correct Optional[str] annotations in kimik2_detector tool call functions
- fix: fix: correct insert_sort logic in add_req_state for ignore_eos scheduling
- [Fix] Honor HiCache timeout with pending pool transfers
- [PD] Stage MLA DCP=8 relayout before Mooncake RDMA for KIMI K3
- Docs
- Python not yet supported