Check out my new in-depth up-to-date training guide here!
Disclaimer
Financial support is very much appreciated, but please do so by donating to my Ko-Fi instead of sending me Buzz, as I currently have no uses for CivitAI Buzz!
Description
An SDXL 1.0 LoRa made by AI_Characters for producing images with the likeness of Emma Watson.
I felt that base SDXL's version of Emma Watson is too overtrained (you can especially see it around the darkened eyes), so I produced my own version of her, and I feel this version has better flexibility than the other Emma Watson models.
Recommended generation parameters
Weight: 1.0 (adjust down as needed when used in conjunction with other LoRa's)
Model: SDXL 1.0 with 0.9 vae
Steps/Sampler/CFG/Resolution: 25 steps | DPM++ 2M Karras | CFG 7 | 1536x1536/1920x1080/1080x1920
'Kohya Deep Shrink' settings: 4 blocks | 2 downscale | 0 start | 0.35 end | skip = true
'CLIPTextEncodeSDXL' ComfyUI node resolution: 4096x4096
Positive prompt: high-definition 8k
Negative prompt: text, blur, low-resolution, low-quality, grainy, glossy, poorly drawn, amateur, poorly shot, compression artifacts
Use "photo, realistic, render" in the negative prompt if prompting the outfit in a different style other than a photo.
Trigger words (must be included in the prompt for the LoRa to take effect!)
'femme ewpp'
Further usage notes
/
Usage permissions
You may not sell this model, merges containing it, or images generated with it.
Credit me when using this model in a merge or hosting it on a generation service.
Changelog
v1.2 'Skynet'
new training config leading to less overtraining, more flexibility and diversity
naming each version from now on after the training config used
for more info see my training guide linked at the top
v1.1
simply switched the token from "ewpp woman" to "femme ewpp", leading to massively better likeness and flexibility
v1.0
Initial release