GPU Coder™ generates and executes optimized CUDA kernels from Simulink® models containing MATLAB® Function blocks. The generated code calls optimized NVIDIA® CUDA libraries, including cuFFT, cuSolver, cuBLAS, cuDNN, and TensorRT. You can use the generated CUDA code within Simulink to perform simulation acceleration, rapid prototyping on all NVIDIA GPU devices, and hardware-in-the-loop testing.
Improve simulation speed by using NVIDIA GPUs.
Generate CUDA code from Simulink models by using GPU Coder.
Build and deploy to NVIDIA GPU boards.
Compare results from model and generated code simulations.
Tune parameters and monitor signals through a TCP/IP communication channel between development computer and target hardware.