Update June 28th, added V3+VAE, pruned and fp16 versions also added.
V3
More training added to supplement some things I feel are missing in current models. Skin textures, darker lighting and more variety in race in people.
V2
After weeks in the making, I have a much improved model. No longer a merge, but additional training added to supplement some things I feel are missing in current models. Lots of new training for skin textures, lighting and non-asian faces to balance out the asian dominance in models. If you create a generic prompt, you'll get a greater variety of races and faces now. Skin textures are increased by a large amount, if that's not your thing, you can put detailed skin in the negative prompt and get back that airbrushed look if you like.
A photorealistic model I've been using for a little bit. I hate smooth airbrushed skin so I refined this model to be very realistic with great skin texture and details. There are 2 models, a standard and an ultra.
The Ultra model is nearly 3 times as large, but it's not 3 times as good. It is better though with a broader knowledge. You can even generate 1024x1024 in many cases. The config file is required for the Ultra model, be sure to download that too.
So I recommend using the normal version unless you have the need or vram to run the Ultra model. If you'd like to run the Ultra model with modest vram, try --medvram or --lowvram in your auto1111 startup script.
A side-by-side comparison. Link to see full-size samples with metadata. Like I said, the Ultra model isn't 3 times as good, but it is better across multiple subjects, not all. The pruned version of the Ultra model is debatable, might be better to use the normal V1 model in some cases. But all have been provided so you can make the choice yourself. For me, I use the Ultra model in every case.
Do you have requests? I've been putting in many more hours lately with this. That's my problem, not yours. But if you'd like to tip me, buy me a beer. Beer encourages me to ignore work and make AI models instead. Tip and make a request. I'll give it a shot if I can. Here at Ko-Fi
Description
Stable Diffusion 1.5 (512 x 512)
A-Zovya Photoreal V1 Ultra Inpainting
VAE recommended vae-ft-mse-840000-ema-pruned.ckpt
FAQ
Comments (15)
V1 ULTRA Inpainting is 5.3 GB. That's the same size as V1 Inpainting model. Is that correct? I'd expect it to be 16+ GB like the base ULTRA model.
yeah, it pruned it in a similar way due to it being based on the inpainting model. But there are slightly different models, so I gave the option anyway.
Hey, where did to 16 gb model go? I can't find it anywhere. </3
click the V1 Ultra tab at the top. Then to the right there's a big blue download button. The full or pruned version will be there depending on your settings. But look just below that to a "files" dropdown and there are more options there.
Looks insane! thanks for your hard work!!!
thanks!
Hello, thank you for your hard work.
When generating a Westerner, realistic skin detail appears, but when trying to generate an Asian male, the realistic depiction drops significantly. If the next version comes out, can I ask for improvement?
I really like your model's skin portrayal T^T
very interesting. thanks for the feedback
Beautiful work. If you don't mind, can i ask "how many references photos you've trained to achieve it ? " I am about to make one myself, gathering 4K-8K references and with some optimization. But i'm not sure "how many" will be enough to get good result for a checkpoint.
You don't need that many. Check a tutorial on training a lora and merge that with a good checkpoint.
@Zovya Thanks. So i assume your final checkpoint wasn't SD 1.5 because of the vanilla SD 1.5 size ? (5.0Gb ish - 7Gb ish )
@Reelai no. most people have left the original SD1.5 behind long ago.
Your UltraInpainting model is god-tier.. it's the best out for photorealism!
thanks!