Transform any image into that signature Gorillaz look—with bold, graphic linework, flat pop-art colors, and a touch of grunge texture. This LoRA is trained on the FLUX.1-Kontext-dev base model and specializes in recreating the iconic, slightly rough-edged cartoon style beloved by the virtual band.
Features
- 🎨 Applies bold, hand-drawn line art & flat color fills
- 🖌️ Adds subtle painterly/grunge textures for that authentic “print-punk” feel
- 👓 Retains original composition & shapes while overlaying the Gorillaz aesthetic
- ⚡ Works seamlessly in Image-to-Image (I2I) workflows on FLUX.1-Kontext-dev
- 🎵 Perfect for album art, avatars, posters, or any “cartoon-punk” transformation
Trigger words
make this image gorillaz styleWhat Works Best
- Portraits & Characters: Human faces, pets, mascots
- Scenery & Objects: Buildings, landscapes, everyday items
- Album/Poster Art: Bold compositions & graphic layouts
- Logos & Icons: Retains readability with a cartoon edge
Description
FAQ
Comments (6)
How did you train the model? It's for an academic paper.😝
I'm interested because of my Disco Elysium model and wanted to know if I should convert images with that style to realistic ones to do some reverse processing in training.
you could do either forward generation or backwards generation for the training data, i did forward generation using kontext max
is this trained on ChatGPT/Sora images? Because the yellow/brownish tint is way too much present.
the training set was generated using Kontext max with a colour palette dominated by muted, earthy tones - deep greens, browns, and golden yellows - with a deliberately aged, sepia-toned quality that gives it an antiquated feel
Amazing lora. Very easy to get desired results. Thanks a lot.