FluxHyperReal

Prompt

A hyper-realistic portrait of a white American man, mid-30s, strong jawline, short dark brown hair, piercing blue eyes, light stubble, and a confident yet approachable expression. His skin has natural pores, slight imperfections, and a subtle sheen from soft lighting. He is wearing a tailored navy-blue suit with a crisp white dress shirt, top button undone. The lighting is cinematic, with a soft key light from the left casting realistic shadows and depth. The background is blurred, featuring a warm, golden-hour urban cityscape with soft bokeh lights. The image is ultra-detailed, 6K resolution, hyperrealistic rendering, with perfect facial symmetry, depth of field, volumetric lighting, and cinematic color grading. The focus is on the subject's face, ensuring an expressive, lifelike result with intricate skin textures and natural reflections.

Prompt

A hyper-realistic portrait of an Indian man, 32 years old, medium brown skin tone with warm undertones, sharp jawline, deep brown eyes, thick well-groomed eyebrows, and short, slightly wavy black hair. He has a neatly trimmed beard and mustache, exuding confidence and intelligence. He is wearing a dark charcoal-grey kurta with a subtle embroidered pattern or a well-tailored navy-blue blazer over a white dress shirt, slightly open at the collar. The lighting is soft and directional, casting natural shadows and highlighting facial contours. The background is a blurred modern cityscape at dusk, with warm bokeh lights, evoking a sophisticated yet grounded atmosphere. Ultra-detailed skin texture with visible pores, fine hair details, and a natural sheen. The image is rendered in 6K ultra-quality, hyperrealistic style, cinematic color grading, volumetric lighting, soft depth of field, and expressive facial details.

Use it with the 🧨 diffusers library

from diffusers import FluxPipeline
import torch

pipe = FluxPipeline.from_pretrained("Himanshu806/FluxHyperReal", torch_dtype=torch.bfloat16).to("cuda")
image = pipeline('A person in a bustling cafe indmodelf').images[0]
image.save("my_image.png")

Please abide by License of Flux Dev model.