from diffusers import AutoPipelineForText2Image
import torch

pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.bfloat16).to('cuda')
pipeline.load_lora_weights('paukraft/pau-lora', weight_name='pau-lora.safetensors')
image = pipeline('[trigger] in his early twenties with light skin and short, slightly messy brown hair. [trigger] is smiling subtly, looking directly at the camera. [trigger] is wearing a dark blue textured knit sweater over a white crew-neck shirt. The background is a soft, neutral beige with even lighting and no shadows, creating a calm and professional studio-like setting. The overall mood is friendly and casual.').images[0]
image.save("my_image.png")

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