( Not my Model )
https://civitai.red/models/941345/hoseki-lustrousmix-anima-v10?modelVersionId=2982979
Anima Base v1
Text encoder and VAE can be found on this HuggingFage page .
Recommended settings :
-
Positive prompt : masterpiece, best quality
-
Negative prompt : worst quality, low quality, jpeg artifacts, signature, sepia, fewer digits, extra digits, bad hands, bad anatomy, watermark, censored, score_1, score_2, score_3
-
Sampler : ER SDE / Euler a - 24 steps
-
CFG Scale : 4-5
I highly recommend using the
ADetailer
extension for better faces and eyes.
The Anima iteration of this checkpoint is a bit of a departure from the Illustrious v2 version, going for a slightly less flat style overall, but keeping a colorful, sharp 2D anime style for faces while getting a little painterly side to bodies and backgrounds, as well as shading.
Just like with other Anima based checkpoints, artist styles work, and natural language to enhance your prompts or add text also works.
Thanks to @ComradeAnanas, @Manityro, and @justTNP for helping create this checkpoint.