This repository provides the official PyTorch implementation of Res2Net, a multi-scale feature extraction backbone for convolutional neural networks published at IEEE TPAMI. The architecture introduces a hierarchical residual-like connection pattern within a single residual block to capture features at multiple scales. Pretrained Res2Net_v1b models are provided, achieving strong performance on ImageNet classification and serving as effective backbones for downstream tasks including object detection, semantic segmentation, and panoptic segmentation.