Krea 2 LoRA — maxmaxer1475/diana-krea2-lora-v1

A DreamBooth-LoRA for Krea 2, trained on Krea 2 RAW and shown on Krea 2 Turbo. The samples below were generated with this LoRA on Turbo (8 steps).

Trigger

Use the token dianafx to invoke the concept.

Samples

"dianafx woman, full body, wearing blue jeans and a white t-shirt, walking through a modern city street, American comic-book illustration"

"dianafx woman, close-up portrait, serious expression, looking directly at the viewer, American comic-book illustration"

"dianafx woman, sitting at a cafe table, wearing an elegant red dress, three-quarter view, American comic-book illustration"

Use it with diffusers

import torch
from diffusers import Krea2Pipeline

pipe = Krea2Pipeline.from_pretrained("krea/Krea-2-Turbo", torch_dtype=torch.bfloat16).to("cuda")
pipe.load_lora_weights("maxmaxer1475/diana-krea2-lora-v1")
image = pipe("dianafx woman, full body, wearing blue jeans and a white t-shirt, walking through a modern city street, American comic-book illustration", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")