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
- [Perf] [AMD] Qwen3-Next: reach the Gluon TP4 AR+norm+quant kernel from the boundary
- [Perf] [AMD] Qwen3-Next: consume the fused AR+norm+quant output
- [Fix] openai: a pre-stream abort answers with its HTTP status, not a 200 stream
- [Muse Spark] Accept top_k=0 as disabled top-k filtering
- [Fix] nixl: release transfer worker when room is aborted mid-wait
- [Fix] qwen3_5: Warn on unknown visual bias in load_weights (#44504)
- [Fix] multimodal: Keep shared-memory feature transport for multi-node pipeline parallelism
- [Fix] multimodal: Preserve input IDs during embedding
- [CI] [AMD] muse-spark: Add a MI35x nightly GSM8K + MMMU accuracy gate
- wip
- Docs
- Python not yet supported