Bringing support for NVIDIA Blackwell architecture across data center and GeForce products, NVIDIA cuDNN 9.7 delivers speedups of up to 84% for FP8 Flash…
Bringing support for NVIDIA Blackwell architecture across data center and GeForce products, NVIDIA cuDNN 9.7 delivers speedups of up to 84% for FP8 Flash Attention operations and optimized GEMM capabilities with advanced fusion support to accelerate deep learning workloads.
Source
Source:: NVIDIA