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:
(NeonMutation: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 achieve the result we may wish to create. Of course to eliminate 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
FAQ
Comments (12)
It would be great in current context if Civitai offered to link multiple models together so to bulk download the entire series in 1 click.
There are some UI's that offer the ability to search and download every model on CivitAI inside the editor, without ever leaving the tool you use. Some people lack the hard drive space also, all my "for mixing" style models are over 100GB even using float16. Once you start merging them together as intended and even merging with your own models too you can quickly reach half a TB. There is an API though to answer your comment but not many people have implemented this as i say. Thanks for checking out these rigid style models, we have more on the way!
AXSTHXTIC.root
While you publish a lot of unique models at once, it would be great if they could be consider as a group nonetheless so to keep things together.
It's a nice idea, you should suggest it to the civitai team, I'm not against people being able to see the full "palette" of style models "for merging" in one place - this essentially gives people more aesthetic "colors" to mix together. The true power is in that gradient form as often the effects are impossible to predict, but extremely powerful!
For now, you can use the "2.1" and "checkpoint" filters, then search for "djz" to find all the original models in this series. Alternatively https://civitai.com/user/AXSTHXTIC.root use that link to get my account. until now this is what i was told to do :) maybe in future there will be new organizational features
When I click your name civitai gives no results. But if I click your link at the first comment you have posted, It works :S I think it is about your nickname have a "dot ." inside and civitai is not indexing your nick due to a bug. For your information. Thanks for the great models.
Considering the size of some files, it might put too much stress on the servers if people downloaded 10s of gigs at a time. Or am I wrong?
I made a post on the discord where I detail the whole thing a lot more.
I suggested contiguous downloads for published bulks which would start a new download only when the previous one is completed.
you guys should look at the civitAI api because there are many people that do not use the website to download the models from here. It's available direct to developers on demand ;) Also load balancing is a very old feature in server infrastructure.
@rockedt also the team have changed my name now so the "." is not longer causing any problems for certain web browsers! thanks!! https://civitai.com/user/AXSTHXTICroot
doesnt work for me im getting an error
remember these are for Stable Diffusion 2.1-768, they do not work in most UI's without copying the correct .yaml settings for inference V. There are many UI's which allow you to run Stable Diffusion, so check with the provider of you generators. I redownloaded this and checked it. It definitely works.
edit: added the default config for this model, however you should check with your webUI or whatever software you use to ensure this it correct.