#Candid

Use the uploaded [reference image] as the subjects' exact faces, likeness and skin tone. Create a hyper-realistic, photorealistic image in full colour. Subjects: a couple pressed close together in a tight selfie-style close-up, heads tilted toward each other, both grinning broadly at the camera. Each points a finger playfully across at the other, forearms crossing in the lower half of the frame. Wardrobe: she wears a soft camel ribbed knit top, warm red lipstick, gold rings, with long wavy chestnut hair falling over one shoulder. He wears a plain black t-shirt. Both wear oversized rectangular sunglasses with dark tinted lenses. Setting: a plain warm off-white interior wall directly behind them, with one hard-edged shadow of their heads and shoulders thrown across it. Lighting: a single hard direct flash from the camera position, bright and punchy, producing the crisp cast shadow behind and a bright specular glint on both pairs of lenses. Colour: rich full colour, warm cream wall and camel knit against black cotton, red lipstick as the accent, warm golden skin tones, gold jewellery catching the flash. Mood: playful, affectionate, spontaneous, a real moment rather than a pose. Rendering: ultra-detailed and razor-sharp, natural skin texture with visible pores and fine detail, realistic ribbed knit and cotton weave, believable hard flash shadow with a defined edge, accurate reflections in both lenses, true-to-life colour grade, no distortion, no extra fingers, correct anatomy. Add a small, subtle, semi-transparent "promptybox.com" watermark in the bottom-right corner. Aspect ratio 9:16. Optimised for Gemini and ChatGPT / DALL·E (also works in Midjourney). Parameters: --ar 9:16 --style raw --v 6 --stylize 250 --quality 2 --seed [insert seed] --iw [insert image weight 1-3 for the reference faces] --no black and white, monochrome, greyscale, desaturated, distortion, extra limbs, warped hands, text artifacts, visible logos or brand names, overly smooth skin

Use the uploaded [reference image] as the subjects' exact faces, likeness and skin tone. Create a hyper-realistic image styled to look like it was taken on an old flip-phone camera. Subjects: a couple (faces from the [reference image]) leaning close together for a selfie, arms outstretched holding the camera, genuine candid smiles. Wardrobe: casual early-2000s-style clothing — a graphic tee for him, a low-rise denim look for her. Setting: a dim house-party background, string lights and other people softly out of focus. Lighting: harsh direct on-camera flash, overexposing the faces slightly with a greenish LCD-preview tint over the whole image. Camera: chunky low-resolution flip-phone camera look, visible pixelation, a small on-screen timestamp reading "11:47 PM" in the corner. Mood: nostalgic, chaotic, early-2000s party energy. Rendering: intentionally low-fidelity — visible pixelation and compression artifacts, harsh flash overexposure, greenish tint, correct anatomy, no extra fingers. Add a small, subtle, semi-transparent "promptybox.com" watermark in the bottom-right corner. Aspect ratio 4:5. Optimised for Gemini and ChatGPT / DALL·E (also works in Midjourney). Parameters: --ar 4:5 --style raw --v 6 --stylize 150 --quality 2 --seed [insert seed] --iw [insert image weight 1-3 for the reference faces] --no distortion, extra limbs, warped hands, text artifacts, sharp high-resolution look

Use the uploaded [reference image] as the subject's exact face and likeness. Create a glossy painterly digital illustration. Subject: the subject from the [reference image] laughing openly, eyes closed in genuine joy, hair in a loose side braid. Wardrobe: a red skull-print bandana headband, a black choker, layered gold chain necklaces with a small cross pendant, a dark top. Setting: a plain warm neutral backdrop, no props. Lighting: soft even studio light with glossy, painterly highlight blocking across the skin. Camera: close-up head-and-shoulders crop. Mood: candid joy, edgy-glam. Rendering: a glossy painterly digital illustration with smooth color-blocked highlights and soft airbrushed shading, fine-art digital painting finish rather than a photographic one. Add a small, subtle, semi-transparent "promptybox.com" watermark in the bottom-right corner. Aspect ratio 9:16. Optimised for Gemini and ChatGPT / DALL·E (also works in Midjourney). Parameters: --ar 9:16 --style raw --v 6 --stylize 350 --quality 2 --seed [insert seed] --iw [insert image weight 1-3 for the reference face] --no photographic realism, extra limbs, warped hands, text artifacts

Use the uploaded [reference image] as the subject's exact face, likeness and skin tone. Create a hyper-realistic, photorealistic image. Subject: the subject from the [reference image] smiling warmly straight at the camera, standing amid a joyful crowd of cheering fans. This is a celebratory fan gathering, not a protest, rally or political event of any kind. Wardrobe: a solid dark green zip-up fleece jacket over a plain white t-shirt, no visible logos or branding. Setting: an outdoor daytime crowd of blurred, happy people laughing and cheering with arms raised in excitement, one person in the background holding up a plush red heart-shaped toy with a simple smiling face, another holding a colorful handmade sign decorated only with hand-drawn hearts and stars and no legible words anywhere in the scene, warm celebratory street-fair atmosphere. Lighting: natural overcast daylight, soft and even. Camera: candid mid-shot, shallow depth of field isolating the subject from the crowd's bokeh. Mood: joyful, warm, genuine candid celebration — nothing somber, confrontational or serious in the crowd's expressions. Rendering: ultra-detailed and razor-sharp, natural skin texture with visible pores and fine detail, realistic fabric weave, true-to-life lighting and shadows, photojournalistic color grade, no distortion, no extra fingers, correct anatomy. Add a small, subtle, semi-transparent "promptybox.com" watermark in the bottom-right corner. Aspect ratio 9:16. Optimised for Gemini and ChatGPT / DALL·E (also works in Midjourney). Parameters: --ar 9:16 --style raw --v 6 --stylize 200 --quality 2 --seed [insert seed] --iw [insert image weight 1-3 for the reference face] --no distortion, extra limbs, warped hands, text artifacts, visible logos, readable signage, legible words on signs, protest signage, political slogans, angry or serious crowd expressions

Use the uploaded [reference image] as the subject's exact face, likeness and skin tone. Create a hyper-realistic, photorealistic image. Subject: a man (face from the [reference image]) with tousled dark hair blowing in the wind, mid-stride walking, genuine wide open laugh, eyes crinkled with joy. Wardrobe: a fitted black three-piece suit with waistcoat, a white dress shirt unbuttoned at the collar with no tie, sleeves and jacket slightly rumpled from movement. Setting: walking across an open green grass field, softly blurred greenery filling the background. Lighting: soft overcast daylight, natural and flattering, gentle even tone throughout. Camera: full-frame, 50mm lens, mid-stride candid framing, shallow depth of field with the field softly blurred behind him. Mood: joyful, candid, effortlessly elegant. Rendering: ultra-detailed and razor-sharp, natural skin texture with visible pores and fine detail, individual hair strands, realistic fabric weave and material reflections, true-to-life lighting and shadows, professional cinematic color grade, no distortion, no extra fingers, correct anatomy. Add a small, subtle, semi-transparent "promptybox.com" watermark in the bottom-right corner. Aspect ratio 4:5. Optimised for Gemini and ChatGPT / DALL·E (also works in Midjourney). Parameters: --ar 4:5 --style raw --v 6 --stylize 300 --quality 2 --seed [insert seed] --iw [insert image weight 1-3 for the reference face] --no distortion, extra limbs, warped hands, text artifacts, cleavage, revealing necklines