Text-conditional image generation model based on OpenAI's unCLIP
Want to make some of these yourself?
