Samaritan 3d Cartoon - CivArchive (CivitAI Archive)

Samaritan 3d Cartoon - v4.0 SDXL

NSFW

Version 4 is for SDXL, for SD 1.5 version please pick version 1,2,3

I don't know a good prompt for this model, feel free to experiment

i also have a lora sdxl for this model
https://civarchive.com/models/121932/samaritan-3d-cartoon-sdxl

For SD 1.5 version make sure to use hi res fix and use a decent VAE or the color will become pale and washed out

if you like the models, please consider supporting me on, i will continue to upload more cool stuffs in the future

patreon.com/PromptSharingSamaritan

https://ko-fi.com/promptsharingsamaritan

Description

Trained for SDXL on 1024x1042 images, overall better hands than in 1.5

FAQ

Comments (31)

monkeyman007Aug 20, 2023· 11 reactions

CivitAI

can you make the training data available to download?

nanunanaAug 20, 2023· 6 reactions

CivitAI

i got the same result without this Lora. Look picture below...

NyoonAug 20, 2023

CivitAI

Anyone know how to use SDXL? I am trying to use this Version 4 XL checkpoint but can't.

I am to understand there are previous steps before I can use it.

PromptSharingSamaritan

Author

Aug 21, 2023

what's your graphic cards

NyoonAug 21, 2023

@PromptSharingSamaritan it's 2080ti, do I need an upgrade? :sad:

kpqneDec 22, 2023

i never did figure it out on AUTO1111, i'll try again, but for now i'm using ComfyUI for SDXL 1.0 and AUTO1111 for 1.5, not sure if i can post a youtube link here or not O_O but ComfyUI takes a bit of geting used to, i just follow tutorials to the letter for a while.

kpqneDec 22, 2023

@Nyoon comfyui seems to "tile" things too large for memory, FYI, but i'm not expert :-D

BizzAIAug 21, 2023

CivitAI

i've never wanted to punch so many images in the face before great work lol it did come out amazing :D

nexys05Aug 23, 2023

CivitAI

Has anyone gotten this (v4) to work on a 3060 with 12GB of VRAM (or GPU with the same VRAM)? I keep on getting "CUDA out of memory" errors.

thesilvermothAug 23, 2023

I have the same GPU. Are you using Automatic 1111? If yes, try --medvram in the webui-user.bat. I can even hires fix to 3x without any issue. It takes a bit longer but it won't fail

rampayj981Aug 24, 2023

Same issue, but with a different SDXL model. Will try using --medvram as mentioned. Our cards don't pack as much punch as they did a year ago 🥲 Technology needs to slow down ffs

nexys05Aug 24, 2023

@thesilvermoth, it works! Thank you for the recommendation :)

thesilvermothAug 24, 2023· 1 reaction

@nexys05 Cool! Glad it helped :D

shakebakentakeAug 28, 2023

Use Low Vram Setting

bccafeAug 31, 2023

New Automatic 1111? 1.6.0 ,try --medvram-sdxl .

shakebakentakeSep 2, 2023· 2 reactions

--medvram-sdxl for AAAA 1.06, this will load medvram only for SDXL models

nexys05Sep 22, 2023

@shakebakentake, many thanks for the tip :)

monkeyman007Sep 2, 2023· 6 reactions

CivitAI

An update that is not overtrained would be cool pls

JustJoeSep 19, 2023

CivitAI

Only checkpoint that wont work for me. Getting error : Error: Could not load the stable-diffusion model! Reason: Error while deserializing header: MetadataIncompleteBuffer

TOMFOOct 11, 2023· 3 reactions

CivitAI

Expression is a little difficult to control, there is always a little frown unhappy feeling, if coupled with a smile, happy word expression will be overly happy, a little exaggerated

tagaevailina892Dec 28, 2023

For which version?

dan000206244Feb 21, 2024· 8 reactions

CivitAI

This model gives you some pretty horrible images, very unrefined. Maybe its my settings, but there are no instructions provided.

PromptSharingSamaritan

Author

Feb 21, 2024

resolution should be at least 1024x1024 for sdxl

50826May 1, 2024

Yeah, there's no information and the person who posted it barely communicated in English. This is a garbage model for sure.

AddictiveFutureMay 2, 2024

Dont read these "smart" guys. Model works well. Everything is generated properly.

aiclotsJul 29, 2024

look at all the successful generations in the gallery below and follow their settings

algisjAug 18, 2024

It works great, real fun. Expressions are a bit hard to control, but feasible A+

bjwgbzpjnxouopkq7844Apr 24, 2024· 9 reactions

CivitAI

It's like youtube clickbait thumbnails and knockoff pixar movie stills were mashed together...

I don't get why people like this model.

ps1copatoMay 29, 2024· 1 reaction

CivitAI

How can I get a sad/crying expression from this? I Already tried the LoRa but it doesn't work properly. Any prompts or advices?

SemorezSep 21, 2024

must use detailer

NourdalJul 9, 2024· 2 reactions

CivitAI

Incredibly funny and stylish 3d cartoon model! I really liked these grotesque facial expressions... :))) Sometimes there are problems with the anatomy, but generally uncritical. Very cool! Thanks!

Checkpoint

SDXL 1.0

by PromptSharingSamaritan

Download (Beta) View on CivitAI