A Stable Diffusion LoRA profile based on Carrie Fisher, aka Princess Leia from Star Wars.

This is v.1 testing a new, quicker method I am experimenting with to create these profiles. V.2 will arrive in a day or two. Already giving great results!

Only using images of Carrie from the first movie trilogy, Obviously I only use high quality images but because of the time period this does like to kick out black and white images maybe 30% of the time, this can be controlled with your prompts.

It is generating really nice pictures already though, so please share your images!

Place it in "stable-diffusion-webui\extensions\sd-webui-additional-networks\models\lora" folder in Automatic1111 WebUI root folder.

Made using kohya-ss.

In txt2img, under "Additional Networks", select "Enable" and under Model 1 select "Carrie Fisher_v1" and assign a weight between 0.5-0.7 (the weight required can change depending on the checkpoint being used - feel free to experiment!). In the prompt text window use "Carrie Fisher_v1" within your prompt to generate images.

Embed has been trained on Stable Diffusion 1.5 model.

I Recommend the model https://civarchive.com/models/4922/grapelikedreamfruit - Gives amazing realistic renders.