https://www.buymeacoffee.com/tripleheadg
This is my DEV model :D
The flux version is a merge of:
FluxWoW https://civarchive.com/models/147933/wowxlpdflux
Tessilation https://civarchive.com/models/801623?modelVersionId=973588
OpenFLux https://huggingface.co/ostris/OpenFLUX.1
Dev-DeDistilled https://huggingface.co/nyanko7/flux-dev-de-distill/tree/main
All the ComfyUI workflows are included in the images and the prompts should be there.
The XL version is a merge of:
Demon Core v1.8 https://civarchive.com/models/155977/demon-core-sfwandnsfw?modelVersionId=177633
WoW_XL v2
7 of 9 v2
And 2 unreleased custom models that I may be releasing soon also. One of them is focused on 2D stuff, like tattoo designs and the other is just something I've been working on for fun.
There were also more than 10 custom models created and merged and deleted due to SSD space constraints in the process.
LoRA used:
https://civarchive.com/models/119403/hdr-scenery
https://civarchive.com/models/126343/better-picture-more-details-lora
https://civarchive.com/models/120774/ahegao-tongue-out-cross-eyed-sdxl-1k
I also trained my own LoRA and merged them too.
I've been working with WarAnakin and Team Crystal Clear and they've taught me a few things about training models which helped form the basis of this new model merge.
Description
FAQ
Comments (12)
Top notch, great job Monkey !
Thank you :D
Heck yeah! Just what I was waiting for! Your checkpoints are fantastic and serve as great merge elements / base checkpoints to work off of. Thanks for what looks to be another great one.
Thank you! I'm glad I can help! Hope you enjoy!
The new version of the model really knocks you down from the very first generation, it's so incredible that it's impossible to believe your eyes. Incredible clarity, detail in the smallest details, well-developed skin, eyes, animal hair, and what kind of nature. A lot of work has been done. I want to express my great gratitude.
I'm really happy you like it! Hope I can continue to make awesome content, that helps people like you make even more awesome content!
i will play with it...
but why "speach" spam every single picture ?
btw if you interessted in tongue out may take a look on my blinking eyes and bit on lips
seams very strange ... much overtrained in faces... i need only 20 steps for more or less good faces, works not good with my lora but jugger yamerl and demon does ;)
@sevenof9247 That is definitely true. I'll have to work on the next version. Honestly this model has not been as good as some of my other models.
By the way. I realize I sort of stole your models name. If you would like, I can change my 7 of 9 models to something else.
In answer to why my good supporter and friend shares the artwork they make with my model it is because they are an awesome person :D Their support has been one of the major driving forces behind my work. As well as the community on the AI revolution discord server owned by Olivio Sarikas, and he himself also :D
Also I found that the clip of the ahegao LoRA, while it did not give my model the ability to generate ahegao faces, caused the quality of generations to improve significantly. I actually didn't use any part of the LoRA which specifically changed the facial expressions or added triggerwords.
@Triple_Headed_Monkey
i see...
all ok 7of9 is not sevenof9 ;)