---
license: mit
---
# SageAttention 2++ Pre-compiled Wheel

🚀 **Ultra-fast attention mechanism with 2-3x speedup over FlashAttention2**

Pre-compiled Python wheel for high-performance GPU inference, optimized for RTX 4090 and CUDA 12.8+.

## 🚀 Quick Installation

### Method 1: Direct Pip Install (Recommended)
```bash
wget https://huggingface.co/ModelsLab/Sage_2_plus_plus_build/resolve/main/sageattention-2.2.0-cp311-cp311-linux_x86_64.whl

pip install sageattention-2.2.0-cp311-cp311-linux_x86_64.whl