Turn boring into WOW!!!! - Model Switch Trick A1111 + Vlad Diffusion

This Model Switch Tricks allows you to use the Style and explresiveness of one Model and transfer it to another Model. Because Models are trained on specific images, they create a style similar to that. But that can be boring at static. With this trick I get the cool style of one model, but then render it again in a realitic photo lock of another model

#### Links from my Video ####

#### Join and Support me ####

[emilianovb]

Thank you Olivio! I love these shorter videos, they are straight to the point and make learning easier.

[darkwing_the_spacecat]

Another day, another Olivio vid that blows my mind! Thank you so much!

[gurilagardnr]

This was so simple and so impactful, once again proving how your insightfulness has become invaluable to many of us.

[Devalocka]

Wow! Olivio! Game changer! Thanks for this one - my whole photorealism pipeline just flipped over. :D

[BrandosLounge]

I've been looking for something like this, thank you very much

[pumanuclei]

Great vid! I have kind of been doing this without actually realising, so the tips and refinement of this technique is much appreciated!

[Tigermania]

I've been doing this recently to solve the problem of boring poses and body shapes using anime models, then using image to image and a realistic model as you have shown. Don't forget in image to image you can set the denoise very low and just send the output from image to image back to image to image and generate again(looping) until you get the desired realism.

[fernandomasotto]

great tip olivio. thank you very much!!

[olbriteinc]

Great tip Olivio!! Simple and beautiful! Awesome

[nicolaspada]

Thank you. Simple, clever and powerful tricks.

[dsin]

The simply and effective tricks are the best. Thanks Olivio.

[TMichael]

I think this method is brilliant! Thanks Olivio!

[siio]

very interesting method. i have to test now ! Thanks.

[michaelbayes]

Thanks Olivio for the great video - I tried this (it worked very well) and then I thought why not create a check point merger with Rev Animated & Realistic Vision with 40% factor you use and I got very similar results in 1 step without the wait to switch models. Maybe others can benefit from this

[fernandosouuza]

Im the only one who likes more the "boring" image cause it looks more natural?

[thegeb]

I've been using this technique as well, although i've been calling it "model bashing". I'd love to know more about clip skip as i'm not sure how it's useful.

[lukmanulhakim]

bruhh, I just found this method 2 days ago and then watch your video, I thought I am the only one use this lol

[Karajorma]

I've done this a few times myself by accident and noticed the improvement but I'd never considered making it part of my workflow.

[Afrmanpeace]

First of all I am using my own checkpoint (Colossus)which is more creative and similar detailed than the other realistic checkpoint's. Then I use sometimes controlnet with reference only. I set it to my prompt is more important. this usually sets the style of my image without changing the prompt. This is very useful for realistic photos. I use highres fix 1.5 with Ultrasharp 4x upscaler. after that I am using image to image to upscale it another 1.5x. after this i am usually using Gigapixel to upscale it further.

[tamizhanVanmam]

I really love it it's very helpful.. like this any way to change the dress alone