Krea 2 LoRA — weyr/shae-face-lora

Prompt

A cinematic wide shot of [character-shae] wearing ornate golden armor, standing atop a floating crystal island amidst a nebula of swirling violet clouds.

Prompt

A cozy, detailed close-up of [character-shae] reading an ancient leather-bound grimoire in a rain-streaked cyberpunk library filled with neon holographic displays.

Prompt

An action-packed scene of [character-shae] diving through a vibrant coral reef, surrounded by bioluminescent jellyfish and shimmering schools of tropical fish.

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 [character-shae] to invoke the concept.

Samples

"A cinematic wide shot of [character-shae] wearing ornate golden armor, standing atop a floating crystal island amidst a nebula of swirling violet clouds."

"A cozy, detailed close-up of [character-shae] reading an ancient leather-bound grimoire in a rain-streaked cyberpunk library filled with neon holographic displays."

"An action-packed scene of [character-shae] diving through a vibrant coral reef, surrounded by bioluminescent jellyfish and shimmering schools of tropical fish."

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("weyr/shae-face-lora")
image = pipe("A cinematic wide shot of [character-shae] wearing ornate golden armor, standing atop a floating crystal island amidst a nebula of swirling violet clouds.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")