4th Feb 2023: V15 added
These "strong style" Models are intended to be merged with each other and any model for Stable Diffusion 2.1
I recommend merging with 0.5 (50/50 blend) then using prompt weighting to control the Aesthetic gradient.
example merged model prompt with automatic1111:
(Apocalypso:1.2) (yourmodeltoken:0.8)
if you drop the "djz" and the "V21" what remains is the token you need to call up the concept in the model. All examples shown were the Raw Token, no other words. Tokens are case sensitive and in almost all models it will match the filename.
It is possible to merge these models with each other using a different value. It is possible to pair models and then merge those resulting models. In this way we can blend abstract concepts together and then weight the tokens to acheive the result we may wish to create. Of course to elminate all those tokens, you can simply train a new custom model from the outputs, which means you are back to a single token.
A video explanation will follow, but for now the above explanation should do. We are focused on getting as many style/aesthetic models into artist hands to enhance the creativity already at their finger tips.
Art Freedom for all!!
[all original artwork used for training with full permission from Drift Johnson]
Description
demo pics no hires
FAQ
Comments (2)
i'm confused, why not merge them yourself then upload that version?
you cannot merge 2.1 and 1.5 together and the datasets are a different resolution. Checkpoint merging offers unlimited interoperability versus other fine tuning methods and thanks to the supermerge extension, take up no more HDD space either. Also, general purpose models cannot be so easily merged due to the problem of overfitting. By using "tight models" or "strong styles" it is possible to explore unique aesthetics, while still affording the use of any Hypernetworks, Textual Inversions and LORA's as you might wish to use them. Examples will be added to the reviews section in time, but check those reviews to see what I mean.
If you are asking why we don't merge our models and upload the merges, there are more than 100 of these style models available. 100x100 = 10,000 and you can merge at more ratios than just 50/50 mixes so if we did that, Civit would just be dominated by our styles and nothing else. We thought it best to only upload the "Primary colors", allowing artists to mix their own shades :)
Lastly, you can only merge so many times before they "melt" or "fall apart". This is why these models are very strong, to allow a couple of generations of merging, Thanks!