The supplementary materials accompanying this work consist of two primary components:

1. Code for Image Classification, Object Detection, and Semantic Segmentation: The code for these tasks is located in their respective directories. Each codebase provides an implementation of TransNeXt in both native PyTorch and CUDA versions.

2. PyTorch CPP CUDA Extension for Sliding Window Path Acceleration: This extension is located in the 'swattention_extension' directory. To utilize the CUDA version of the TransNeXt implementation, one must first navigate to the 'swattention_extension' directory and install the CUDA extension using the command 'pip install .'. Please ensure that the CUDA version on your machine matches the CUDA version of PyTorch during installation. 