✦ Arthemy Comics Illustrious - Quick Start ✦
I really struggled to generate "western illustrations" the Illustrious model and after a lot of tries and more than 300 iterations and 60 LoRA trained on my own handmade illustrations and handwritten prompts, I think I've managed to create a pretty reliable solution.
I suggest you to use my quick prompts as a "quick start" and then move from there in the directions you prefere.
Afterall - they are called Checkpoints for a reason!
1_____Size: Around 1024 x 1280
2_____Prompt: Start from the prompts present here and tune the keywords based on your needs:
Quick Prompt
masterwork, masterpiece, cartoon screenshot, movement, depth, complex lighting, lineart, complex shadows, character design, old-school, vintage fantasy, dynamic pose, official art, (insert your prompt here!)
Quick Negative Prompt
sketch, chibi, big head, collage, low quality, worst quality, text, signature, jpeg artifacts, bad ???????, bad hands, six fingers, extra fingers, heterochromia, blurry, graphite \(medium\), yellowed, muted, (no lineart), (3d:0.7), limited palette, (insert your prompt here)
3_____CFG: 5.0 to 7.0
4_____Steps: 45 to 50
(I've noticed that 30-40 steps are not usually enough to solve the image completely)
I hope you like my models and I can't wait to see what you all are going to create with it!
____________________________________
Merged / subtracted / inspired by Models:
Nova Furry - https://civitai.com/models/503815/nova-furry-xl
Toonify - https://civitai.com/models/36281/toonify
Ri-Mix - https://civitai.com/models/99??95/ri-mix-pony-illustrious
Velvet Paradox - https://civitai.com/models/1734917
IllustraVersa - AdV3nt - https://civitai.com/models/1888661/illustraversa?modelVersionId=2233553
UltraComix - https://civitai.com/models/1798149?modelVersionId=2034957
Unholy Desire Mix - https://civitai.com/models/1494558?modelVersionId=2200584
+ A lot of my own trained LoRA
If you like Arthemy Comics, remember to check these other models and consider to support their creators!
Description
The objectives with this version were:
+ More Creativity: the heavy use of small LoRAs in my first version collapsed most of its generations into the same shapes. Now, with the help of many personally trained LoRA I've been able to push the creativity of this model further and have more variation of characters and poses.
+ More Flexibility: It's still a heavily comicbook inspired model, but if you write for example "Realistic" it will provide a more realistic comicbook style and so on. So, now you can explore different aesthetic with the same model - while keeping it "Comicbooky"!
- Less reliable hands: Unfortunately, training this model on cartoons and fantasy illustrations compromised its ability to provide five fingers consistently - I think it can be easily fixed with a LoRA.
This is going to be the last version for a while, since I'm quite satisfied with the results and, unless your feedback will make me change my mind, I think I'll take a small break.
Have fun! I can't wait to see what you're going to create with this model!
FAQ
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.