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('Vedantk1301/JHNAB_Flux_LoRA_v2', weight_name='JHNAB_Flux_LoRA_v2.safetensors')
image = pipeline('JHNAB stands against a pitch black seamless backdrop, the surface subtly textured to catch the gentlest glimmers of light. He is lit by a single narrow spot key from above and slightly to camera left, creating dramatic chiaroscuro that sculpts his cheekbones and beard. A faint rim light from behind on camera right separates him from the void. JHNAB wears a sharply tailored black leather moto jacket over a matte black crewneck tee. His pose is upright, shoulders squared, head tilted just 5° toward the key light. His expression is intense and probing, brows slightly furrowed and lips parted in a thoughtful exhale. Use the provided reference images to capture his exact facial structure, hairline, and beard density. Render in a crisp editorial style, 4:5 vertical crop, f/2.8 shallow depth to blur the backdrop, ISO 100, 1/200s shutter.').images[0]
image.save("my_image.png")

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