Advoluptuous

218

244

#Anime

#femelle

#coloré

#vivid

#Modèle de base

#filles

#smooth

#sharp

#basemodel

Advoluptuous

This model is a merge of a variety of anime models (70% personal models, 20% anime/ models, 10% other), fine-tuned using Dall-E 3 and Midjourney data. Designed to produce creative and sometimes surprising output without being too wild.

How to gen

The look you will get will wildly depend on what VAE you use. I recommend the standard kl-f8-anime2.ckpt VAE if you want vibrant results. Otherwise, you may want to use a blessed-type VAE with lowered contrast.

Using hires fix is a must. For detail, ADetailer extension isn't essential, but it will make a big difference for your gens.

For the upscaler I use 4x-UltraSharp, but other upscalers work fine too. Latent upscaler may give you unpredictable results.

This model will work with any image ratios, but for some more extreme ones it may produce multiple panels or character-sheet-like output. You can often fix this by specifying background details in the model.

The model is highly responsive to negative prompts, so if you see unexpected output, that will do the job. For example you may like to use “border” in neg prompt if the model adds borders to your gens.

Works well with SFW, content, and to some extent abstract and landscape-type art.

Get free BUZZ

If you like the model and produce some cool output with it, I'll tip you 100 Buzz on your review. If you make lots of amazing images I'll tip you up to 400 Buzz.