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] diffusion/attention: Fuse VSA tile packing and layout conversion
- [Perf] [AMD] GLM5.3: Allow AITER MXFP4 scatter intermediate only for GLM-5.3-Flash on ROCm
- [Fix] hicache: Evict pass-through Mamba host checkpoints before path tails
- [Fix] distributed: Keep deprecated parallel_state getters traceable under torch.compile
- [Perf] sampling: Avoid full-prompt copies in custom thinking budget logit processor
- [Perf] disaggregation: Send replicated KV heads from one prefill rank
- [Fix] [AMD] deterministic: Run block-FP8 linears through Triton on ROCm
- [Fix] metrics: Correct MFU/MBU estimator overcounting
- [Bug] benchmark: --dataset-name random silently requires the ShareGPT download; fails on offline/restricted networks
- [Perf] [AMD] gdn: ReplaySSM verify uses one pipeline stage at every batch size on gfx950
- Docs
- Python not yet supported