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
- [Feat] [Intel] lora: Add intel_xpu LoRA backend
- [Fix] scheduler: Wait for idle before completing abort-mode pause
- [Fix] hicache: Support flat pages for asymmetric K/V host pools (head_dim != v_head_dim)
- [Perf] hicache: Optional parallel page reads in the file storage backend
- [Perf] [NVIDIA] diffusion: Support per-token shift/scale in the fused norm_scale_shift kernel
- [Fix] function_call: Accept more Harmony call shapes in GptOssDetector
- [Perf] mistral: Resume the compact-format JSON scan while streaming (34 KB args: 2.6 s to 43 ms)
- [RFC / Feature] SPICY-KV: 28.6x–50x Pre-RoPE Algebraic Multigrid KV Compression for RadixAttention & Long-CoT (91.7%–100% RULER Recall & 100% Shared-Prefix Multi-Query Reuse)
- [Fix] lang/runtime: Raise HTTP errors in async generation
- [Perf] [NVIDIA] fp8-gemm: L2-aware CTA swizzle for the SM121 (GB10) blockwise FP8 GEMM
- Docs
- Python not yet supported