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
- [Doc] blogs: Sync LMSYS SGLang blog cards
- [Perf] [NVIDIA] diffusion/joy: Enable packed-image QK normalization and RoPE on SM103
- [Perf] [AMD] kimi-k3: Fold the latent MoE RMSNorm into up_proj
- [Chore] server-args: Drop stale references to removed cuda-graph flags
- [Feat] hicache: Expose drain and L3 backup failure metrics
- [Feat] kda: run the Kimi-K3 fused FlashInfer decode on a BF16 SSM state (NVIDIA SM100/103)
- [Perf] diffusion/ming: Fuse lossless RMSNorm scale and residual epilogues
- [Feat] [CPU] model: Support Phi-4-W4A16-AutoRound model on CPU
- [Doc] [Apple] apple_metal: Make pyproject swap safe to re-run
- [Perf] expert_pack: Deduplicate route_ids and use .tolist() in ExpertPackStore.acquire
- Docs
- Python not yet supported