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
- [Feat] rust-server: Give the native /generate hook the request, its headers and its outcome
- [Feat] sgl-router: Render MiniMax-M3 and M3.1 through sglang-processor
- [Refactor] [NPU] graph: Remove the NPU graph runner
- [Fix] forward-hooks: Warn when CUDA graphs skip Python hooks
- [Fix] [AMD] dsv41: Make Engram's input_ids optional for the AMD monokernel
- [Fix] function_call: Preserve valid DeepSeek calls when a sibling is malformed
- [Perf] Capture GDN prefill in breakable CUDA graphs
- [Perf] hisparse: Use deque for HiSparseCoordinator.ack_staging_queue FIFO drain
- [Fix] profiler: Ignore PREBUILT batches in profile V2 stage mapping
- [Test] minimax-m3.1: Add nightly e2e tests on H200 and B300
- Docs
- Python not yet supported