NOTE : Base model : Z Image Turbo & Flux1.
Showcase images were created with ComfyUI and are in PNG, include metadata & workflow.
VERY IMPORTANT NOTE :Do not like my models, do not give me buzz, do not follow me, otherwise you will be reported to Civitai for attempted corruption
Description
The improvements concern the following points:
A finer and more detailed drawing
A more complex and scenographic composition
Tendency oriented more towards realism than anime
More subtle rendering of light
Recommended tag: Niji Style.
This model shows some flexibility, it is not limited to Niji style, you can test other type of rendering, like anime or cinematic photography
FAQ
Comments (5)
I have had amazing result with this LoRA. Thank you!
Just Amazing! Thank you - it's fantastic.
A Stan Katayama version for ZiT or Qwen 2512 would be awesome. Thanks!
I'm actually planning to finalize a Niji LoRA for Z Image
@Adel_AI Wow man, you really did it. I've been looking for something that would bring Katayama's Niji 6 sdxl model (mainly mj niji 5 expressive trained) to a modern model and this looks to have nailed it even more than the Flux version, and by a lot. Thanks so much for this.