(Not the band...)
I used the ComfyRoll Model Merge Stack to create this model!
Merged AnimeGodXL with AnImagenXL and ChikokomenXL, plus a couple of personal lora and also some sliders.
I don't cherry pick results, I don't inpaint either. not even High Res Fix was used to create these images. Just default comfyui workflow.
https://www.buymeacoffee.com/tripleheadg
https://freedevproject.org/faipl-1.0-sd/
As per the description of AnimagineXL, if you use this model in a merge or modify it in some way you must provide the link to this license agreement.
This is, to date, one of, if not the best anime model available for SDXL. Capable of a variety of styles, from realism to CGI, to pure anime and lineart/manga style artwork. Depending on the positive and negative prompts that you use it can be a highly effective model. And even with simple prompts it still produces imagery that most other anime models would strive for with a full paragraph of prompting.
Anime lovers unite!
Models used to create this:
WoW_XL v2
7 of 9 v4 Joybringer
LoveXL v2
Niji SE
The Talos Project
XL6 Hephaistos
SimpleDiffusionXL
SpectrumBlendV2
Yamer's Anime Ultra Infinity v3
Samaritan 3D
7 of 9 V2
WoW_XL v1
RunDiffusionXL beta
NightVisionXL
Yamer's Merge Heaven
SDXL Base Model
Lora:
https://civarchive.com/models/158413/alice-in-wonderland-sdxl
https://civarchive.com/models/134606/pe-big-mom-character-one-piece
Shout out's to all the creators of these models, go try them all out for yourselves as each one has it's own unique style! All worthy of praise for different reasons and purposes.
Prompt:
Album Cover Art Son Goku from Dragonball Z going super-sayajin 1, keyframe anime screencap, SSJ1 (in the style of Akira Toriyama), Album Cover Art, often for music albums, band branding, or iconic cover designs.
--
belted A detailed oil painting of Monkey D. Luffy from One Piece set in a vintage port city. The atmosphere is dark and intricate, featuring elements of biopunk and horror. The color palette focuses on dark azure and red, adding a mystical and slightly psychedelic touch to the landscape.
Description
The fith iteration of my custom re-merged over model-sets.
FAQ
Comments (15)
U should merge a few related LORAs into the mix just to add soem variety
100%
My plan is to first create a merged LoRA and then merge that with the model. At this point in time I do not have any anime LoRA though.
@Triple_Headed_Monkey Be careful. For my XL prompts LORA have done more harm than good. I prefer the style prompts and get better results.
@dappergenius491 If I do merge LoRA I will try to do so indirectly, perhaps by merging the LoRA with the base model and then merging with the base model. Not tried this yet, but I think it could compensate for some of the major LoRA issues.
you are deceiving people with these photos and the prompts you have in the description. you are untrust worthy and should be banned.
This is not a nice way of interacting with people. I will provide evidence of the workflow now. And then you can do what you want with it.
This is for ComfyUI.
If you need any assistance in setting it up to check the advocacy of my claims, Just shoot me a message. Despite getting off on the wrong foot, I'm more than willing to help you improve your outputs across the board.
https://drive.google.com/drive/folders/15gT1weovQjESLpX7533eQrZyy4eWsrwB?usp=sharing
Here you go.
Here is the image, regenerated today. Using the same exact workflow I used to generate all of the images. If there is a particular image you would like to know more about, just ask.
I mean really. If I was trying to deceive people I could easily have done a better job of the artwork than this....
@Triple_Headed_Monkey I heard the punch from here
If you are having problems with this model, maybe you should give up on Stable Diffusion. This and JuggernautXL are the best models I have seen on CivitAI. The current top models for favorites and reviews don't come close to following prompts as well as this and JugernautXL. I use Invoke AI and an absolutley floored at what I can do with this model.
@dappergenius491 So glad to hear that you're enjoying the model! Makes all the sleepless nights in recent times worth it!
If you have problems with the model, just ask for help. Please don't make this kind of accusation without a shred of evidence.
I guess it is a compliment of a sort that you think the output of the model is so good that you'd accused Triple_Headed_Monkey of faking them 😂
@Triple_Headed_Monkey You handled this really well 😊
@NobodyButMeow Thanks. I don't want to get angry at people for my own problems. Technically he's only angry with me because I haven't put the effort in to make the generation information available, which is something I intend to start doing with all my uploads from now on!
I can't seem to get this to load in Automatic1111 1.6. I can load the base SDXL model as well as all others I've tried like JuggernautXL, nightvisionXL... etc...
Any ideas?
It has the fixed fp16 VAE baked in. Perhaps that is why?
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.