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
- [Docs] Add Ascend A3 deployment to the Hy3 cookbook
- [Qwen3.8-Next] Add pipeline-parallel PD prefill with MTP
- [Qwen3.8-Next] Fuse PLE hash and pinned-host embedding gather
- Pin remaining unpinned H2D copies in ForwardBatch init hot paths
- [Benchmark] Fix generated-prefix cache reuse across orderings
- [Bug][ROCm] PP + DSA: HIP fault on multi-chunk prefill; failure threshold tracks --cuda-graph-backend-decode
- [Bug][ROCm] --cuda-graph-backend-decode breakable fails at startup on DSA models: 'NoneType' object has no attribute 'get_page_table_64'
- [Bug][ROCm] --dcp-size > 1 crashes every rank on the first forward: forward_absorb_rocm_core unpacks 2 values from a 1-value return
- [Bug] enable_aiter_allreduce_fusion auto-enable is commented out while the log still claims it was enabled
- [Fix] Size routed expert capture buffers for speculative verification
- Docs
- Python not yet supported