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 aiter/wave cuda_graph_custom_mask undersize on long-context spec verify
- fix: replace hardcoded layer_id with mtp_start_layer_idx in Step3p5 MTP
- Redact secret fields from server_info responses
- fix: use system DP rank for prefill sender
- perf: use torch.empty for HiCache dummy read buffers to skip a memset
- Build SWA window kv buffers for the EAGLE draft-extend cuda-graph path
- docs: Added pending support warning for DiffusionGemma Cookbook
- [Do Not Merge][BCG][2/N] Enable DSV4 prefill BCG: two-pass capture + DSV4-aware mem reserve
- Fix stale decode offload state on host allocation failure
- Fix duplicate FPM binding across attention CP ranks
- Docs
- Python not yet supported