Waifu Diffusion is a latent text-to-image diffusion model that has been conditioned on high-quality anime images through fine-tuning.The model originally used for fine-tuning is Stable Diffusion 1.4, which is a latent image diffusion model trained on LAION2B-en. The current model has been fine-tuned with a learning rate of 5.0e-6 for 10 epochs on 680k anime-styled images.<a href="https://gist.github.com/harubaru/f727cedacae336d1f7877c4bbe2196e1" rel="ugc" target="_blank">See here for an in-depth overview of Waifu Diffusion 1.3</a>.Team Members and AcknowledgementsThis project would not have been possible without the incredible work by the <a href="https://ommer-lab.com/" rel="ugc" target="_blank">CompVis Researchers</a>.<ul><li><a href="https://github.com/harubaru" rel="ugc" target="_blank">Anthony Mercurio</a></li><li><a href="https://github.com/sALTaccount/" rel="ugc" target="_blank">Salt</a></li><li><a href="https://twitter.com/cafeai_labs" rel="ugc" target="_blank">Cafe</a></li></ul>In order to reach us, you can join our <a href="https://discord.gg/touhouai" rel="ugc" target="_blank">Discord server</a>.<a href="https://huggingface.co/hakurei/waifu-diffusion-v1-3" rel="ugc" target="_blank">Originally posted to HuggingFace</a>

Waifu Diffusion

This version supports most of the popular <a href="https://danbooru.donmai.us/wiki_pages/tag_groups" rel="ugc" target="_blank">danbooru tags</a>. Check out this <a href="https://gist.github.com/harubaru/f727cedacae336d1f7877c4bbe2196e1" rel="ugc" target="_blank">post for examples</a>.

Files with SHA256: 4afab9126057859b34d13d6207d90221d0b017b7580469ea70cee37757a29edd