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
- [Fix] utils: Keep next_power_of_2 callable from Triton JIT kernels (#48460)
- [Fix] benchmark: Honor max_completion_tokens in OpenAI datasets
- [Feat] hicache: Expose host-tier occupancy and write-back blocking metrics
- [Perf] [ROCM] diffusion: Ring attention on aiter, and reuse the ring's K/V buffers
- [Doc] diffusion: fix nonexistent ImageSamplingParams in support_new_models guide
- [Feat] comfyui: Add an offline workflow converter and reject weight patches on SGLD models
- [Feat] [NPU] HiCache: Support Mooncake L3 for DeepSeek-V4 and GLM-5.2
- [Perf] reject_sampling: Vocabulary-tiled final-token pass at small batch
- [CI][NPU] CI: Add GLM-5.3-w8a8 1P1D GPQA case and release data pipeline
- [Perf] dflash: Side-stream context projection and a one-kernel verify commit
- Docs
- Python not yet supported