AI RESEARCH
Evaluating CUDA Tile for AI Workloads on Hopper and Blackwell GPUs
arXiv CS.LG
•
ArXi:2604.23466v1 Announce Type: new NVIDIA's CUDA Tile (CuTile) Our results show that CuTile effectiveness is strongly workload- and architecture-dependent. On datacenter-class Blackwell (B200), CuTile achieves up to 1007 TFLOP/s for fused attention, outperforming FlashAttention-2 by 2.5x while requiring only 60 lines of Python kernel code. For GEMM, CuTile reaches 52-79% of cuBLAS performance in 22 lines of code (versus 123 for WMMA), making it a practical replacement for hand-written CUDA kernels but not yet for vendor-optimized libraries.