MATLAB Coder Support Package for PyTorch and LiteRT Models

Generate C/C++ code from PyTorch and LiteRT

With MATLAB Coder or Simulink Coder, the MATLAB Coder Support Package for PyTorch and LiteRT Models enables you to generate generic, target-independent C/C++ code for PyTorch and LiteRT models. You can deploy a variety of pretrained deep learning networks, including YOLOv11, Whisper, DINOv2, Depth Anything, and SAM2. You can generate optimized code for pre-processing and post-processing along with the trained neural network, enabling deployment of complete applications. Code replacement libraries can be used to incorporate processor-specific intrinsics for target hardware (e.g., ARM Cortex-A/M processors).

When used with GPU Coder, you can generate plain CUDA code for PyTorch and LiteRT models along with your MATLAB code and Simulink models. The generated CUDA code can be deployed to NVIDIA Jetson and DRIVE platforms using the MATLAB Coder Support Package for NVIDIA Jetson and NVIDIA DRIVE Platforms.

Additionally, when used in Simulink with Simulink Coder, you can simulate Simulink models that incorporate PyTorch and LiteRT models.

This support package is available for R2026a and later releases.

If you encounter download or installation problems, please contact Technical Support.