Conditional GAN (cGAN) is a variation of Generative Adversarial Networks (GANs) where both the Generator and Discriminator are conditioned on additional information, such as labels or attributes. This conditioning allows the model to generate outputs that are aligned with specific conditions, such as creating images of certain objects or styles. By incorporating external information, cGANs enhance the model's ability to produce targeted and relevant data, making them useful in applications like controlled image synthesis and data augmentation.