WD 1.5 Beta is fine-tuned directly from stable-diffusion-2-1 (768), using v-prediction and variable aspect bucketing (maximum pixel area of 896x896) with real life and anime images*.* Given the broad range of concepts encompassed in WD 1.5, we expect it to serve as an ideal candidate for further fine-tuning, LoRA's, and other embedding applications.
It is important to note that this is a preliminary release of the model and that we are committed to improving the model as time progresses. Developing a model that can handle a wide range of styles and concepts requires a substantial investment of time and computational resources.
Given that this model has undergone extensive fine-tuning from SD 2.1, it may generate responses that deviate from models to which you are accustomed. To optimize your experience, we recommend that you take some time to familiarize yourself with WD 1.5.
For best performance, you must use the WD VAE with WD 1.5. The VAE can be found here: https://huggingface.co/hakurei/waifu-diffusion-v1-4/blob/main/vae/kl-f8-anime2.ckpt
Example Positive Prompt (Anime):
1girl, aqua eyes, baseball cap, blonde hair, closed mouth, earrings, green background, hat, hoop earrings, jewelry, looking at viewer, shirt, short hair, simple background, solo, upper body, yellow shirt,
(waifu, anime, exceptional, best aesthetic, new, newest, best quality, masterpiece, extremely detailed:1.2)
Example Positive Prompt (Real Life):
beautiful, maid outfit, 1girl, detailed hair, portrait, simple background, a girl with long hair and cherry blossoms,
(waifu, realistic, real life, exceptional, best aesthetic, new, newest, best quality, masterpiece:1.2)
Example Negative Prompt:
lowres, ((bad anatomy)), ((bad hands)), text, missing finger, extra digits, fewer digits, blurry, ((mutated hands and fingers)), (poorly drawn face), ((mutation)), ((deformed face)), (ugly), ((bad proportions)), ((extra limbs)), extra face, (double head), (extra head), ((extra feet)), monster, logo, cropped, worst quality, jpeg, humpbacked, long body, long neck, ((jpeg artifacts)), deleted, old, oldest, ((censored)), ((bad aesthetic)), (mosaic censoring, bar censor, blur censor)
images generated with the aesthetic model