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
- [multimodal] Stop packing per-image features only to split them again
- [multimodal] Stop each per-item feature shipping the whole packed buffer over IPC
- [multimodal] Ragged feature storage, and a per-image split for Step3
- [parser] Hold back harmony markers split across streaming chunks
- [HiCache] Fix Mooncake registration for split MLA host buffers
- [NPU] [bugfix] Fix ascend sampling TypeError with min_p.
- Offload OpenAI prompt prep (template render + tokenize) off the event loop to fix streaming ITL stalls
- Foundry SGLang integration
- [Draft] [CPU] Deterministic inference on Xeon platform
- [Kernel][DS V4.1] Add W8A8 block-FP8 GEMM configs for H20-3e (DS-V4.1, 32x32)
- Docs
- Python not yet supported