One-Shot Fusion
Details
Download Files
Model description
This merge is a mix of One-Step Illustrious and One-Step Realism. All example images were generated without loras or upscaling, and only used some automatic inpainting during generation.
Though I mostly tested this model in the default style, it seems particularly good at switching between default, anime, and realistic styles (see prompts section for how). If you're looking for an even stronger realistic or anime style, consider one of the other models in the One-Step Collection.
DISCLAIMER
While this model can generate SFW images, it is NSFW and capable of generating explicit material. Use caution in your prompts if you're looking to avoid anything NSFW, but I can't guarantee a random NSFW image won't appear from time to time.
RECOMMENDED SETTINGS
These are the settings I use, but there could be something better. Feel free to experiment on your own.
Resolutions: 832x1216 / 768x1344 / 1024x1536 / 1248x1824
The 1248x1824 resolution is surprisingly stable compared to One-Step Illustrious and One-Step Realism. Most of the example images were generated at that resolution. You might still come across occasional body proportion issues though, so if you want something a bit more stable, use one of the lower resolutions.
Steps: ~35
CFG Scale: 7
Sampler: Euler Ancestral
Scheduler: Karras
Upscaler / Hires. fix (Optional): 0.3 strength, 2x upscale, 4x-Ultrasharp or SwinIR_4x
ADetailer / CLIPSeg (Optional): Use as desired/needed
RECOMMENDED BASE PROMPTS
Same as above, this is just what I use. Feel free to experiment to your own preference.
Positive: masterpiece, best quality, amazing quality, detailed eyes
Negative: bad quality, worst quality, worst detail, sketch, censor, bar censor, extra fingers, extra toes
For a flat anime style, add "anime coloring" to positive and "realistic, photorealistic" to negative. For a more realistic style, add "realistic, photorealistic" to positive. If you're trying to generate an image without a visible face, be sure to change "detailed eyes" to "detailed" in the positive prompt.



















