Type | |
Stats | 298 661 |
Reviews | (28) |
Published | Aug 31, 2023 |
Base Model | |
Training | Epochs: 4 |
Usage Tips | Clip Skip: 1 |
Hash | AutoV2 DB74496E38 |
A custom trained/tuned version of SimpleXL with more dynamic poses and interesting compositions.
Tis a very strange model, my first attempt at a full model. My intentions with this were to try and keep the results as similar to the original Simple_XL as possible, whilst also making things more interesting overall. It's sort of a cross between realism and artistic styles. With the intent being to produce imagery that allows both realistic looking subjects and at the same time allows it more artistic freedom over how that realism is presented in less realistic fashions.
Support me here!
https://www.buymeacoffee.com/tripleheadg
At 1024x1024, and similar resolutions at different aspect ratios, it produces good results. However best results are seen by further using it to refine the upscaled image. Somewhat like the hi-res Fix feature of Automatic1111. Almost all defects that may occur in anatomy (eyes/fingers) are corrected simply through this additional step.
It is more than capable of any style, much like most SDXL models with the right prompting. Considering how well it follows the prompt anyway, I doubt it will be much of an issue for anyone.
It does nudity, but the model is focused on the more artistic magical side of things. It's not supposed to generate NSFW content and has not been trained to do so, or with that intent.
Images generated using the Neg4all LORA @ 0.25 strength
I'm currently working on training a full VFX model, however I need to learn a tonne more, so watch this space and hopefully we'll get some interesting updates for the future!