StudioGAN provides implementations of 7 GAN architectures, 9 conditioning methods, and multiple regularization modules for training and evaluating generative models. The library includes a comprehensive benchmark suite with 8 evaluation metrics (FID, IS, PRDC, IFID) and supports standard datasets (CIFAR10, ImageNet, AFHQv2, FFHQ). It also compares GAN performance against auto-regressive models (MaskGIT, RQ-Transformer) and diffusion models (LSGM++, CLD-SGM, ADM-G-U).