A LoRA for our lovely Master Tigress!

V1

The Lora is trained on 30 selected images of Tigress. Training includes NSFW and SFW work, so both output types are possible.

2 Epochs at 1500 Samples each at 768x768 with bucketing.

Suggested use:

I suggest the weights stay between 0.5 and 0.8 for best results (see example images).

The trigger word is "master_tigress".

Other helpful prompts include:

• detailed red eyes

• yellow sclera

V2

Trained on more images and with a larger resolution.

Suggested use:

I suggest the weights stay between 0.3 and 0.7 for best results (see example images).

The trigger word is "master_tigress".

Other helpful prompts include:

• detailed red eyes

• yellow sclera

To get here usual outfit, you may use

• chinese clothing, vest

Known Issues:

The model really wants to output NSFW images, except if prompted negatively against it.
When you get non-perfect eyes, copy the image to inpaint and let it run on just the eyes again.

描述:

First version

训练词语: master_tigress