READ BEFORE DOWNLOADING
This model has some artifacting, but this can be kept low by generating with over 40 steps, and by not using a DPM-based sampler at lower step counts. It's most apparent when generating animals, but does still sometimes appear on human faces. Higher step counts usually help.
Don't use this model expecting a balanced general art experience. It's weird, finicky, and bugged, but it would've been a shame to let its specific strengths go to waste as well.
This unstable experiment model based on 526Mix V1.2 has a few strengths:
High-detail portraits
Close-up shots
Interiors
Semi-realism
Sci-Fi
It also has issues:
The aforementioned artifacts (semi-realistic faces will often need fixing)
Landscapes are usually messier and sometimes not as expressive
Some art styles aren't as good (e.g. line art, Jack Davis style)
Input
Per image generation will cost 0.0047$
For premium plan image generation will cost 0.00$ i.e Free.
Output
Unknown content type
Related Models
Discover similar models you might be interested in
Lin Yuchun / Hayashi I-Jun - v2.0-HOT
Lin Yuchun / Hayashi I-Jun - v3.0-ERO
Nene Kusanagi - SD 1.5
[Clothing] Hair Bow Variation / - side_of_head-v1.0
QR Code M - v2.0
Futanari-Style revic - v1re
Yuuta Kadowaki | Otonari no Tenshi-sama ni Itsunomanika Dame Ningen ni Sareteita Ken / The Angel Nex



![[Clothing] Hair Bow Variation / - side_of_head-v1.0](https://images.modelslab.com/?Image=https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/68d7f848-bcc8-4bb5-bd6d-18a0751d7206/width=1024/3557748.jpeg)





