TIGER Framework: A New Breakthrough in GPU-Accelerated Fully Homomorphic Encryption for Large Model Inference

Introduction / Main Floor: TIGER Framework: A New Breakthrough in GPU-Accelerated Fully Homomorphic Encryption for Large Model Inference

This article introduces TIGER, the first GPU-accelerated high-precision TFHE fully homomorphic encryption framework. Through programmable bootstrapping and batch processing design, it achieves order-of-magnitude acceleration on key non-linear layers such as GELU, Softmax, and LayerNorm, providing a feasible solution for privacy-preserving cloud deployment of large models.