The 50 images below are hereby released into the public domain as-is and are therefore free to use or modify in any project, paid or free, without permission or attribution.
You can download the full collection with the button below or use right-click "Save as..." on specific images below.
These images were generated by a VQGAN AI model trained on a fine-tuned version of OpenAI's 512x512 class-conditional ImageNet diffusion model and use CLIP to generate images from text prompts generated with this script. The text prompt used is listed beneath each image.
