santa hat
deerdeer nosedeer glow
Sign In

FoolKat_Anime_Merged_Model

176
1.4k
19
Updated: Oct 5, 2024
base modelanime2d2.5dflat art
Verified:
SafeTensor
Type
Checkpoint Merge
Stats
405
Reviews
Published
Jul 6, 2023
Base Model
SD 1.5
Usage Tips
Clip Skip: 2
Hash
AutoV2
30B7DAD7AA

7/6/23 Update - I decided to create a version 3 and end the series with version 3. Version three adds one more model to the mix, my favorite anime model, DucHaiten-GODofSIMP.

7/5/23 Update - I have created a Lora that achieves the style I'd been attempting with this merge and it works well when combined with this model. The lora is called Anime Lineart Style. As a result, I won't be creating a version 3 to try to achieve the style any longer. Version 2 will most likely be the final version of this merged model.

Version 2.0 - Still haven't been able to get the specific style I wanted, so I may have to create a style lora and merge that into the model for version 3. I did get a flat 2D anime style for this version, which contrasts the 2.5D images version 1.0 created.

Version 1.0 - This is a merged model of three different models that I really like in an attempt to get a specific style of anime artwork. Unfortunately, I failed to get the results I was hoping for. I did, however, succeed in making a pretty decent model for 2.5D anime-style images. I'm going to try to create a different mix of the models to see if I can get the results I was actually going for, and if it works I will publish it as a version 2.

The current model does SFW well, but has issues with NSFW. It works well for nude portraits, but most of my attempts to create sex scenes resulted in strange errors like detached body parts, extra limbs, etc.

Model was created by merging BokoMix2, LottaMix, and 3moon_anime_line models.

This model doesn't require a VAE. I baked in Anything4.0 VAE.

All of the images I upload for this model are unedited and don't use any special extensions like ControlNet. I do not do any inpainting or img2img to correct errors. Why? Because you deserve to see how the model works without any touchup or special extensions like ControlNet. You can surely get better images by putting in the extra work, which I do for my personal images. The samples are just meant to showcase what the model can do raw with nothing but prompt engineering.