If you use it, please share your results so I can be aware of how should I tweak for next version
This model is an experiment for allowing high quality of mature content, is in early stages, but soon enough will be producing next version, so far I recomend using 0.5, depending on how much you want to see clothes, it's kinda flexible so you might be able to do many prompts without issues.
In version 0.3.5 it's more balanced, if you use strength 1, it's more likely to get ????? depending the model/checkpoint, and with 0.5 you can have ?????? images not that explicit but you should experiment yourself to find the best combination
Description
A less agressive version, trained with other dataset I gathered and more flexible for the descriptions you might want to use but I'm 75% sure that is SFW most of the times without explicit descriptions and depending on how strong you use the model, although every 0.1 can make some difference and 0.2 makes a more visible change in the resulting image. Not what I wanted but still gives interestings results.
/me -researching -creating datasets
FAQ
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.