VQGAN + CLIP is our first steps into computer vision via Generative Adversarial Networks.

These experiments were made using Python and 3x Nvidia 3090 GPUs.

The AI shown below generates trippy videos from text prompts.