This LoRA was inspired by an Image made with SD3.5
All the prompts I used are included in the example images to get you going!
Trained on: Flux.dev fp8
Images made with: Flux.dev fp8
Sampling method: Euler
Sampling steps: 30
Distilled CFG Scale: 3.5 but you can experiment as well
Clipskip: 1
I am using Hires.fix with the following settings:
Upscaler: 4x_NMKD-Superscale-SP_178000_G or 4x-AnimeSharp
Hires steps: 10
Denoise: 0.20
Upscale by: 1.2x - 4.0x
This LoRA was inspired by an Image made with SD3.5
All the prompts I used are included in the example images to get you going!
Trained on: Flux.dev fp8
Images made with: Flux.dev fp8
Sampling method: Euler
Sampling steps: 30
Distilled CFG Scale: 3.5 but you can experiment as well
Clipskip: 1
I am using Hires.fix with the following settings:
Upscaler: 4x_NMKD-Superscale-SP_178000_G or 4x-AnimeSharp
Hires steps: 10
Denoise: 0.20
Upscale by: 1.2x - 4.0x