This can be said to be the model I spent the most time, effort and money on, after a lot of training and the last time was 65340 steps, I finally produced the official product. 30% pixar + 20% 3D realism + 50% 3D anime, it ended up getting close to what I wanted. The most impressive thing is always the eyes, which I always focus on the most, followed by the leather and other things, hope that all of that can create a new sense of excitement for everyone.
If you like my content, support me at:
https://linktr.ee/Duc_Haiten
For those who don't have a powerful enough GPU, or are just too lazy to install enough complex software to run the model. mage.space will be a great site for you, with the number of models increasing every month, including my model of course. Especially for only $15 per month you can have unlimited photo creation for yourself, no credit score, no time, absolutely truly unlimited photo creation in a month.
Nprompt:
illustration, 3d, 2d, painting, cartoons, sketch, (worst quality:1.9), (low quality:1.9), (normal quality:1.9), lowres, bad anatomy, bad hands, vaginas in breasts, ((monochrome)), ((grayscale)), collapsed eyeshadow, multiple eyebrow, (cropped), oversaturated, extra limb, missing limbs, deformed hands, long neck, long body, imperfect, (bad hands), signature, watermark, username, artist name, conjoined fingers, deformed fingers, ugly eyes, imperfect eyes, skewed eyes, unnatural face, unnatural body, error, bad image, bad photo
Description
FAQ
Comments (11)
This is pretty good mate! Tx!
Damn Duc, this is incredible looking. You've done a lot of great stuff, but this might be the one I'm most looking forward to trying out.
hope it won't let you down
Once you press the download button, there is no turning back. （〃｀ 3′〃）
Can't reproduce any of these. What am I doing wrong? :D
you must have done something wrong, if you want the exact same you need to install everything exactly like my picture. Not only negative prompt and prompt, you need seed number, step count, CFG number, hires fix parameters
@DucHaiten Thanks for the response. I think I figured it out. My install was a bit borked, re-installed everything and now I am able to reproduce the images :)
Hey, and thank you for all your cool models. Have been using them a lot.
I am unable to get this one to work, even though I have copied it exactly. Here you can see my settings and the result I get:
https://imgur.com/a/ohaQ5M0
The only difference I can see is the name of the actual model. This is what I get when I copy all settings from your picture:
Steps: 20, CFG scale: 10, Sampler: DPM++ 2M Karras, Seed: 500664419, Size: 704x960, Model: pixar-beta-v12, Model hash: c55f672c1b, Hires steps: 10, Hires upscale: 2, Hires upscaler: R-ESRGAN 4x+, Denoising strength: 0.5
Here the model is called pixar-beta-v12, while I am using duchaitenGoldenlife_v10. Is this a different version of the model or what am I getting wrong here?
no difference at all, I changed the name when posting the model to civitai. check again vae and clip skip. In addition, depending on the platform you are using, some hidden settings such as using seeding from CPU or GPU also have an impact.
@DucHaiten Thank you so much for the reply. Which VAE should I have and which clip skip should I use? Not using any atm, as I didn't see them in your setting.
@Illuvatar If I remember correctly, I used clip 1, but clip only changes the image without causing errors. Usually, clip 2 helps the model understand the prompt better. VAE does not need to be added because it is already included in the model, just make sure not to add any additional VAE. Also, try lowering the CFG and hires fix or changing to another Sampler. Sometimes, the issue might be due to differences in your system or platform causing the error.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.