from diffusers import AutoPipelineForText2Image
import torch

device = "cuda" if torch.cuda.is_available() else "cpu"

pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.bfloat16).to(device)
pipeline.load_lora_weights('Keltezaa/candice-swanepoel-2010s-flux-lora', weight_name='candicesflx1.safetensors')
image = pipeline('Create a hyper-realistic portrait of a regal woman inspired by the "Game of Thrones" aesthetic, but dressed in battle-ready attire. She stands in a windswept mountain pass, her armor gleaming under the muted sunlight. Her outfit combines elegance with practicality: a fitted leather cuirass adorned with intricate carvings of wolves and stars, layered over chainmail. A dark cloak with a fur-trimmed collar billows behind her, hinting at her noble lineage. Her blonde hair is tied back in a loose braid, with small strands framing her strong, determined face. She carries a finely crafted longsword at her side, its hilt encrusted with subtle gemstones. Her piercing gaze reflects both beauty and unyielding strength. The scene is detailed with rugged terrain, distant snow-capped peaks, and the faint howl of the wind, evoking an air of mystery and power').images[0]

For more details, including weighting, merging and fusing LoRAs, check the documentation on loading LoRAs in diffusers