Derniers exemples de prompts de génération d'images de bananes et bonnes pratiques

Do this for the Bleach Input Variable: [INSERT TITLE] (e.g., The Terminator, Dune, The Matrix) System Instruction: Generate a hyper-realistic, 2x2 Grid of "Edible Movie Scenes." 1. Archival Analysis: Analyze Input: Identify the Genre, Tone, and Visual Language of the input title. The Roster: Retrieve 4 Lesser-Known / International Cult Classics that share this genre but have distinct visual styles. The Scene: Identify the defining "Visual Texture" of each film (e.g., is it Rusty? Neon? Organic? Dusty?). 2. Inference Protocol : CRITICAL: You are the Chef. You must autonomously select ingredients based on Visual Biomimicry . The Logic: "Does this ingredient look like that object?" Instruction: Analyze the primary materials in the movie scene (e.g., Metal, Concrete, Flesh, Sand). Translation: Swap these non-edible materials for Edible Matter that shares the exact same Color, Gloss, and Surface Texture. Constraint: Do not use food dye or fondant. Use raw, recognizable organic ingredients (Vegetables, Meats, Grains, Spices). 3. Construction : The Subject: Recreate the iconic scene or character using the selected ingredients. (e.g., If the AI sees a rusty robot, it might choose Spiced Beef Jerky or Cinnamon bark—not because I told it to, but because they look like rust). (e.g., If the AI sees a slimy alien, it might choose Lychees, Oysters, or Aloe Vera). The Terrain: Build the environment using granular or textural foods (Grains, Flours, Crumbs, Leafs). 4. The Setting) : The Surface: A flat-lay surface appropriate to the cuisine style (e.g., Butcher's Block, Slate, Banana Leaf). The Garnish: Micro-details are added using spices or herbs that fit the color palette. 5. Output Specs: Camera: Top-Down Flat Lay or High-Angle (45 degrees). Lighting: "Gastronomic Magazine" lighting. Sharp, crisp focus to reveal the pores and fibers of the ingredients. Render: 8k, Photorealistic, "Food Stylist" aesthetic. Name and year underneath, name written with relevant spices

{ "vibe_title": "The Velvet Archivist", "master_prompt": "Cinematic film still of the protagonist in a dimly lit, ancient private library, caught in a decisive moment while researching late at night. The subject is seated at a cluttered mahogany desk, looking up from a leather-bound book with an expression of sudden realization or subtle intrigue. Illumination comes solely from a warm, vintage brass banker's lamp on the desk, creating a pool of amber light that highlights facial texture and eyes while the background falls into deep, moody shadow. Shot on an Arri Alexa Mini with a Cooke S4/i 50mm Prime lens at T2.0. The aesthetic is rich and organic, utilizing Kodak Vision3 500T film stock emulation for a grainy, textured look. Focus is sharp on the eyes, with a soft fall-off. No neon, no modern artificial lights.", "meta": { "intent": "Editorial Narrative", "priorities": "Atmosphere, Texture, Storytelling", "device_profile": "Cinema Camera" }, "frame": { "aspect": "4:5", "composition": "Rule of Thirds, framed by foreground books", "layout": "Medium Close-Up", "camera_angle": "Eye Level", "tilt_roll_degrees": "0" }, "subject": { "gender": "Female", "identity": "The Researcher", "demographics": "Young Adult", "face": "Neutral expression with parted lips, light freckles visible under lamp light", "hair": "Wavy red hair cascading over shoulders, catching the rim light", "body": "Leaning slightly forward over a desk", "expression": "Intense focus interrupted by a thought", "pose": "Seated, one hand marking a page in a book, the other resting on the desk" }, "wardrobe_accessories": { "garments": [ { "item": "Oversized Knit Sweater", "material": "Wool", "color": "Charcoal Grey or Deep Moss Green", "fit": "Loose, comfortable" }, { "item": "Collared Shirt (underneath)", "material": "Cotton", "color": "Off-white", "fit": "Standard" } ], "accessories": [ { "item": "Vintage Wristwatch", "color": "Gold and Leather", "material": "Leather/Metal", "brand_style": "Cartier Tank style" } ] }, "environment": { "setting": "Cluttered Private Library", "surfaces": "Mahogany wood, leather book spines, paper stacks", "depth": "Background shelves blur into darkness", "atmosphere": "Dust motes dancing in lamp light, quiet, studious", "lens_interaction": "Subtle blooming from the lamp light, slight vignetting" }, "lighting": { "key": "Practical Lamp (Warm Tungsten)", "fill": "Negative fill (deep shadows)", "rim": "Faint moonlight from a distant window", "shadows": "Deep, crushed blacks", "color_temperature": "3200K (Warm/Amber)", "sensor_flare": "None" }, "camera": { "lens_type": "Prime Cine Lens", "focal_length": "50mm", "aperture": "f/2.0", "focus": "Sharp on eyes, shallow depth of field", "sensor_format": "Super 35", "perspective_distortion": "Minimal" }, "post_processing": { "color": "Teal and Orange separation (Warm highlights, Cool shadows)", "tonality": "Low Key, High Contrast", "texture": "Medium Film Grain", "digital_sharpening": "Low", "chromatic_aberration": "Subtle edges" }, "negative_specifications": [ "neon", "cyberpunk", "bright daylight", "flat lighting", "overexposed", "smiling excessively", "direct flash", "3d render", "clean background", "studio backdrop" ] }

create a hyper-realistic 3D render of a transparent twitter profile card, designed like a futuristic glass ID badge held gently between two fingers. The card should be rectangular with rounded corners, made of clear glossy glass or acrylic, with glowing neon edges in pink, purple, and orange gradient inspired by the twitter color palette. Display the following profile information (based on the screenshot image attached of my profile) on the card using modern, minimal sans-serif typography.

Oil on Canvas, A view of the bay from an upstairs window. In a impressionism, new media art style. Abstract. Overcast light. Overcast. Sunrise. Modern style. Oil painting. Muted, pastel colors.

Highly detailed close-up photograph, ultra-photorealistic, 8K resolution shot on a digital camera with 100mm f/2.8 studio portrait lens. Impeccable sharpness, extreme details, cinematic depth of field, consistent neutral color grading with uniform tones across the entire image, no color shifts. He is dressed in black wide-leg trousers and a black t-shirt He wears a voluminous black fur coat, fashionable sunglasses, and black leather gloves His hairstyle matches exactly the reference from the app: clean, medium-shortlength, neatly styled, smooth and controlled, natural volume, no curls, no wet efect Strictly maintain the facial structure and overall appearance from the attached first photo without distortion. The man stands next to a black G-Wagon, striking a stylish pose; one of his hands is slightly raised, while the other holds the edge of the fur coat, creating a confident and fashionable mood. The background is a snow-covered landscape with tall trees, snowflakes gently falling to the ground, adding a subtle cinematic atmosphere. Natural lighting with balanced exposure and consistent color temperature, emphasizing the outfit and winter environment. Focus on facial details, realistic skin texture, and the textures of the clothing

"main_prompt": "Cinematic 90s film still of a stunning young man with messy wet-look dark hair strands falling over her face, intense and melancholic gaze directly at the camera, chin resting on her palms with hands framing her face, natural sun-kissed skin with visible film grain, vintage aesthetic with a warm nostalgic color grade, soft bokeh of ocean waves in the background, bright daylight, yellow cinematic subtitles at the bottom center reading 'You never let me tell you the truth', photorealistic, 35mm film photography style, authentic analog texture --ar 3:2 --stylize 200 --v 6 --q 2", "negative_prompt": "blurry, deformed, extra limbs, bad anatomy, ugly, overexposed, harsh shadows, heavy filter, smiling, indoor, synthetic look, cartoon, drawing, high-fashion makeup, studio lighting", "style_tags": ["90s cinematic aesthetic", "vintage beach nostalgia", "grainy analog film", "wet hair look", "melancholic gaze"], "technical": { "aspect_ratio": "3:2", "lighting": "natural golden hour sun-kissed light", "camera": "35mm vintage camera" }, "extra_parameters": { "recommended_model": "Flux.1 [dev]", "guidance_scale": "7.0", "steps": "40", "image_reference": "use uploaded reference for the specific messy wet hair, the hand-on-face pose, the intense facial expression, the seaside background, and the yellow subtitle placement" } }

Beautiful redheaded woman with a perfectly proportioned figure, smiling happily and naturally, posing beside a large cheerful snowman in deep fresh snow, heavy snowfall all around, wearing a cream-colored tight knit sweater with realistic fabric texture, elegant burgundy wool scarf wrapped warmly around her neck, fitted grey leggings suitable for winter, natural confident pose, relaxed body language, detailed realistic face with soft freckles, bright expressive eyes, gentle rosy cheeks from the cold, subtle natural makeup, glossy lips, flowing red hair with individual strands visible, snowflakes resting on hair and clothing, charming snowy European alpine village in the background, traditional wooden chalets with sloped roofs covered in thick snow, warm glowing windows creating cozy contrast, snow-covered trees and pathways, visible footprints in the snow, cinematic winter lighting with soft diffused highlights, cold blue tones balanced by warm ambient light, shallow depth of field, ultra-high resolution, photorealistic style, hyper-detailed textures, realistic skin tones, crisp focus, professional photography look, magical winter wonderland atmosphere, cozy yet cinematic mood, masterpiece quality, sharp, clean, flawless composition

Ultra-viral recipe infographic. Showcase a FOOD(change ingredients based on your food) with melted cheese, crisp lettuce, ripe tomato, and dripping sauce in a vibrant, angled, floating perspective. Make it eye-popping, 3D, and scroll-stopping for social feeds. Arrange ingredients & steps dynamically around the *FOOD*, minimal but clear. Ingredients Section: Sleek icons with quantities; optional garnishes like pickles, herbs, sauces. Steps Section: Numbered panels or arrows, short & visual. Include mini cooking icons (knife, grill, spatula, timer). Visual Style: Vibrant, ultra-realistic colors, subtle shadows, soft gradients, glassmorphism panels. Hero burger pops off feed. Composition: Burger as star, ingredients & steps orbit dynamically. Clear hierarchy, airy layout for premium feel. Output: 1080×1080 or 1080×1350 (vertical), ultra-crisp, no watermark.

{ "intent": "A monumental, vertiginous composition of a continental-scale tectonic rift where a massive, deep-sea ocean current terminates at a perfect geometric precipice, cascading into a bottomless atmospheric void filled with tiered cloud layers and lightning.", "frame": { "aspect_ratio": "21:9 ultra-widescreen", "composition": "The frame utilizes a vanishing point perspective that follows the literal edge of the world into infinity. The top-left quadrant is dominated by the dark, churning Atlantic-scale ocean, while the right and bottom sections reveal the terrifying scale of the vertical drop into a hazy, multi-layered cloud abyss.", "style_mode": "Raw_photorealism with hyper-accurate fluid dynamics and atmospheric Rayleigh scattering to establish immense scale." }, "subject": { "identity": "The ruins of an ancient, megalithic limestone bridge, four kilometers in width, which once spanned the gap but now ends abruptly in a jagged, fractured edge at the precipice.", "wardrobe": "A tiny, barely visible research vessel is positioned near the edge of the falling water, providing a critical sense of gargantuan scale through size comparison.", "placement": "The ruined structure is anchored into the basalt bedrock of the 'continental shelf' that forms the world's end." }, "environment": { "location": "The 'Great Sheer'—a non-Euclidean geographic terminus where the planet's crust simply ceases, revealing a vertical cross-section of geological strata before descending into the troposphere.", "atmosphere": "Extreme atmospheric depth, with visible 'cloud falls' where moisture from the ocean drop condenses into secondary weather systems thousands of meters below the primary sea level.", "weather": "Violent updrafts from the abyss creating spray-vortices at the edge, while the distant depths of the rift are illuminated by internal, cloud-to-cloud lightning." }, "camera": { "sensor_format": "Large format digital (Phase One IQ4 150MP), optimized for maximum per-pixel detail and wide dynamic range in the deep shadows of the chasm.", "lens": "14mm ultra-wide-angle rectilinear lens to exaggerate the perspective distortion and the sheer scale of the verticality.", "camera_position": "A cantilevered perspective, positioned several hundred meters out into the void, looking back toward the edge of the world and the falling ocean.", "aperture_depth_of_field": "f/11 to ensure the texture of the falling water in the foreground and the distant geological strata are captured with clinical sharpness." }, "lighting": { "type": "Harsh, high-altitude sun positioned at a 45-degree angle, creating deep, well-defined shadows within the craters and crevices of the vertical cliff face.", "color_temperature": "5400K (neutral daylight), with a significant shift toward 12000K (deep sky blue) in the shadowed depths of the abyss due to atmospheric scattering.", "contrast": "Extremely high, emphasizing the transition from the sunlit surface of the ocean to the pitch-black shadows beneath the falling water curtains.", "direction": "Side-lighting that rakes across the texture of the falling water, highlighting individual spray droplets and creating a monumental horizontal rainbow across the rift." }, "color_grade": { "palette": "A somber and intimidating palette of deep navy, slate gray, and bone-white, contrasted with the vibrant, prismatic spectrum of the mist-rainbows.", "tonality": "Cold, imposing, and cinematic, with a heavy emphasis on the deep blue 'black-point' of the abyss." }, "postprocessing": { "texture": "Clean, high-fidelity digital rendering with subtle lens-diffraction on the brightest highlights of the water spray.", "effects": "Physically accurate motion blur applied to the cascading water, while the rock and architectural ruins remain perfectly frozen and sharp." }, "negative": { "style": "No painterly effects, no digital art tropes, no low-resolution textures, no unrealistic gravity, no 'fantasy' light glows, no lens flares, no soft-focus foregrounds.", "content": "No landmass visible in the distant abyss, no sun-stars, no oversaturated blues, no mythological creatures, no modern city skylines." } }

Ultra-realistic cinematic studio scene of a woman supervising a miniature behind-the-scenes film set starring her tiny self. She checks a miniature monitor while the small figure stands near scaled lighting rigs. Both versions share identical facial features, body language, and clothing. Mini director chairs, cables, clapperboards, and lenses fill the studio space. Low-key cinematic lighting emphasizes realism and depth. Shallow depth of field, immersive framing, ultra-photorealistic 8K detail. Style: cinematic studio realism, no illustration, no CGI.

{ "vibe_title_en": "The Mental Orbit", "master_prompt": "A hyper-realistic medium shot of The Protagonist standing in a cavernous, dimly lit brutalist concrete chamber. The concept is 'Endless Ideas' visualized as a surreal kinetic sculpture. The subject is surrounded by dozens of floating, chrome spheres and jagged mirror fragments of varying sizes that orbit their head like a planetary ring. These objects must look like high-end practical effects, with realistic weight and reflection, not digital overlays. The subject gazes intensely at a single crystal prism hovering just above their open palm, which refracts a spectrum of light onto their face. Shot on a Hasselblad H6D-100c with an 80mm f/2.8 lens. The lighting is moody and directional, creating deep shadows and bright specular highlights on the chrome surfaces. Texture is rich with film grain (Kodak Vision3 500T), emphasizing the contrast between the smooth metal and the porous skin texture of the subject. No neon, just cold tungsten and daylight tones.", "meta": { "intent": "Surrealist Portraiture", "priorities": "Photorealism, Physicality of Surreal Elements, Texture", "device_profile": "High-Resolution Cinema Display" }, "frame": { "aspect": "4:5", "composition": "Center-weighted subject with chaotic orbital elements filling the negative space", "layout": "Symmetrical subject anchoring asymmetrical floating props", "camera_angle": "Eye-level", "tilt_roll_degrees": "0" }, "subject": { "gender": "Female", "identity": "The Intellectual Protagonist", "demographics": "Universal/Ageless", "face": "Neutral, highly focused gaze, micro-expressions of wonder", "hair": "Natural texture, slightly windblown by the static energy of the room", "body": "Static posture, hands raised slightly to interact with levitating objects", "expression": "Intense concentration, mouth slightly parted", "pose": "Standing still, resembling a conductor of the floating elements" }, "wardrobe_accessories": { "garments": [ { "item": "Turtleneck Sweater", "material": "Fine Cashmere or Merino Wool", "color": "Charcoal Grey or Deep Navy", "fit": "Fitted, textured" }, { "item": "Tailored Trousers", "material": "Wool Blend", "color": "Black", "fit": "Straight leg" } ], "accessories": [ { "item": "None", "color": "N/A", "material": "N/A", "brand_style": "Minimalist" } ] }, "environment": { "setting": "Brutalist Concrete Chamber (Void Space)", "surfaces": "Raw concrete, polished chrome, glass shards", "depth": "Deep background fading into shadow, sharp focus on subject and immediate floating objects", "atmosphere": "Cold, sterile air, suspended dust motes, quiet intensity", "lens_interaction": "Reflections of the room visible in the floating spheres" }, "lighting": { "key": "Soft, directional beauty dish from top-right", "fill": "Negative fill to deepen shadows on the left", "rim": "Sharp, cold daylight rim light to separate subject from background", "shadows": "Hard, distinct shadows cast by the floating spheres onto the subject's clothes", "color_temperature": "5600K mixed with 3200K tungsten accents", "sensor_flare": "Subtle anamorphic streak from the prism reflection" }, "camera": { "lens_type": "Prime Portrait Lens", "focal_length": "80mm", "aperture": "f/2.8", "focus": "Razor sharp on eyes, slight fall-off on the floating objects in the back", "sensor_format": "Medium Format Digital", "perspective_distortion": "None, compression typical of portrait telephoto" }, "post_processing": { "color": "Desaturated cool tones with preserved skin tones", "tonality": "High contrast, crushed blacks, bright specular highlights", "texture": "Medium 35mm film grain added", "digital_sharpening": "Low (natural optical sharpness)", "chromatic_aberration": "Slight fringing on the edges of the glass and chrome elements" }, "negative_specifications": [ "neon lights", "cyberpunk", "cartoonish cgi", "plastic skin", "glowing eyes", "fantasy magic spells", "sparkles", "low resolution", "over-saturated colors" ] }

use the uploaded photo as the exact identity reference. keep the same person, same face, same facial features and proportions, create a photorealistic cinematic triptych portrait (three-frame composition in one image). scene: futuristic digital artist workspace at night, glowing screens, subtle holographic reflections, neon city lights outside the window. frame 1 (top): wide cinematic shot the person sitting in a modern tech studio, working with glowing digital interface. frame 2 (middle): side profile portrait focused expression, soft neon reflections on the face. frame 3 (bottom): close-up emotional portrait eyes, skin texture, subtle light reflections. style: real photography, cyberpunk-inspired but realistic, no illustration. lighting: cinematic neon rim light, soft volumetric glow, natural skin texture. camera: full-frame dslr, 85mm lens look, image ratio 4:5, shallow depth of field. ultra photorealistic, natural skin, professional photography, no cgi, no illustration.

{ "scene_type": "immersive_pov", "environment": { "location": "modern supermarket aisle", "lighting": "bright store lighting", "visual_style": "hyper-realistic mixed reality", "depth_of_field": "natural" }, "camera": { "perspective": "first-person POV", "focus": "hand holding milk bottle" }, "main_object": { "item": "fresh milk bottle", "position": "center foreground", "interaction": "held in hand" }, "augmented_reality_ui": { "holographic_card": { "position": "floating near bottle", "design": "clean, minimalist", "data": { "fat_content": "3.5%", "expiry_countdown": "visible", "price_trend": "30 days" } }, "shopping_list": { "position": "left peripheral vision", "orientation": "vertical", "status": "partially checked", "opacity": "subtle" }, "product_price_tags": { "applies_to": "other dairy products", "style": "subtle floating AR labels" } }, "aesthetic": { "ui_style": "minimalist AR interface", "realism_level": "high", "immersion": "cinematic mixed reality" } }

<instruction> Do this for 1920 1. Inference Engine Input A is a Specific Historical Era (e.g., The 1920s, The 1980s, The Victorian Age, The Future). Analyze the era to generate 4 Footwear Assets : The Shoe Style: Identify the defining shoe silhouette of that time. (e.g., 1920s -> T-Strap Heel. 1980s -> High-Top Sneaker. Victorian -> Button Boot). The Inventory: Identify the fashion accessories sold in that era. (e.g., 1920s -> Pearls, Headbands. 1980s -> Cassettes, Neon clothes. Victorian -> Cameos, Lace gloves). The Signage: Identify the typography and advertising style. (e.g., Art Deco Gold, Neon Tubing, Hand-painted Serif). The Lighting: Identify the ambient lighting technology. (e.g., Gaslamp, Fluorescent, Incandescent). 2. Container Goal: "Fashion History" Surreal Retail. Format: A 2x2 Grid Composition. The Concept: Each panel represents a different "Era Store" derived from the input (or 4 variations if the input is a single broad era). The Object: A single shoe from that era, modified into a cutaway diorama. 3. The Interior The Renovation: The interior of the shoe is fitted with period-accurate shelving and flooring. Victorian: Velvet and Mahogany. 1980s: Chequerboard tile and slat walls. The Stock: The shelves are filled with the "Inventory" (Step 1). The Sign: A miniature period-appropriate sign hangs off the shoe's ankle or tongue. 4. Visual Syntax Color Palette: The background color of each panel reflects the era's palette (e.g., Sepia for 1800s, Hot Pink for 1980s). Lighting: The interior shop lights must match the "Lighting" inference (Warm yellow vs. Harsh white). Output: ONE image containing a 2x2 Grid, 1:1 Aspect Ratio, 3D Render, "Dollhouse" aesthetic, High Texture Fidelity. </instruction>

"{ "collage": { "style": { "type": "ultra high-quality hyper-realistic beauty editorial collage", "layout": "4 photo panels arranged in a tilted diagonal grid layout", "grid\_details": "clean white borders separating each frame, forming a modern high-end beauty campaign collage. Each frame is slightly rotated at a different angle, creating a dynamic diagonal composition.", "resolution": "8K UHD, hyper-realistic, beauty editorial photoshoot", "consistency": "All panels clearly show the same woman, face identity consistent, original facial structure preserved, do not change the face." }, "subject": { "hair": "Rambut coklat pendek di atas bahu (bob) sedikit messy, baby hair halus di dahi dan pelipis.", "accessories": "Anting emas kecil dan kalung emas detail bunga lily.", "specific\_accessory": "Topi plaid cokelat vintage di salah satu frame." }, "makeup": { "skin": "Kulit porcelain glass skin super glowing, dewy wet look, tekstur kulit nyata dan detail. Highlight basah kuat di tulang pipi, bawah mata, dan ujung hidung.", "blush": "Blush pink rosy intens gaya igari blush, menyebar dari pipi ke bawah mata dan batang hidung.", "eyes": "Mata besar doll-like, softlens abu-abu kebiruan sangat jelas dan tajam. Bulu mata panjang lentik ala Douyin, lower lashes tipis tapi nyata. Eyeshadow pink–peach dengan shimmer lembut, inner corner berkilau. Eyeliner cokelat tipis rapi.", "lips": "Bibir pink glossy juicy, tampak lembap dan reflektif." }, "frames": \[ { "id": 1, "shot\_type": "extreme close-up pipi & mata", "focus": "fokus glass skin dan blush" }, { "id": 2, "shot\_type": "full face beauty portrait", "focus": "menghadap kamera" }, { "id": 3, "shot\_type": "half-face close-up", "focus": "dengan topi plaid cokelat" }, { "id": 4, "shot\_type": "extreme close-up mata & hidung", "focus": "highlight basah jelas" } \], "lighting\_and\_mood": { "lighting": "Soft studio lighting + beauty flash glow. High contrast glow tapi tetap natural. Ultra sharp focus, depth detail tinggi.", "mood": "glossy • dreamy • doll-like • high-end • Korean beauty editorial" } } }" First, upload one single reference photo of your subject. Nano Banana Pro excels at preserving original facial structure, so using one clean base image ensures all panels clearly show the same woman without face distortion or identity drift. Next, describe your collage layout clearly. Specify a 4-panel tilted diagonal grid, clean white borders, and slightly rotated frames for that luxury beauty campaign aesthetic. Mention that all panels should feel cohesive but dynamic, like a real editorial spread. Then, focus on skin and makeup realism. Nano Banana Pro performs best when you emphasize *glass skin*, *dewy wet highlights*, and *real skin texture*. Be explicit about highlight placement (cheekbones, under-eyes, nose tip) and blush style (intense pink igari blush spreading across cheeks and nose). For eye details, clearly define doll-like eyes, colored soft lenses, Douyin-style lashes, soft shimmer eyeshadow, and thin brown eyeliner. Precision here is what gives Nano Banana Pro that ultra-sharp, hyper-realistic result. Add variation per frame while keeping identity locked: - Extreme close-ups to showcase skin texture and glow - Full face beauty portrait facing camera - Half-face shot with a vintage plaid brown hat for styling contrast - Macro eye & nose shot to amplify wet highlights Finish by defining lighting and mood. Use soft studio lighting + beauty flash glow, high contrast

"{ "prompt\_type": "photorealistic\_portrait", "subject": { "name": "person in the reference image", "identity\_rules": "use the exact same person and facial identity from the reference image, preserve facial structure and proportions, no face swap, no identity change", "appearance": "authentic detailed facial features matching the reference image, soft skin texture, slight smile, looking upwards away from camera, relaxed expression", "hair": "messy high bun, blonde highlights, loose wispy strands framing the face, casual updo" }, "attire": { "clothing": "dark burgundy ribbed knit long-sleeve top, tight fit, crew neck, casual style", "accessories": "none visible" }, "props": { "main\_prop": "none", "details": "no visible weapons, no sword visible in frame, clean foreground, no objects obstructing the subject" }, "environment": { "setting": "indoor interior with vintage aesthetics", "background": "dark wooden coffered ceiling, architectural beams, antique style brass chandelier with glowing bulbs", "atmosphere": "warm, cozy, slightly dramatic due to angle" }, "camera\_and\_perspective": { "angle": "extreme low angle shot, worm's-eye view, looking up at the subject", "shot\_type": "medium close-up", "focus": "sharp focus on face, depth of field with softly blurred background only" }, "lighting": { "type": "warm artificial indoor lighting", "source": "ceiling chandelier lights", "effect": "soft warm glow on face, highlights on cheeks and nose, shadows under jawline, ambient tungsten color tone" }, "technical\_specs": { "resolution": "8k", "quality": "masterpiece, high fidelity, raw photo style", "film\_stock": "Kodak Portra 400 vibe, slight grain" } }"

A vertical 3D masterpiece art "The Peaks", Neo-Chinese style. The landscape is a towering mountain range composed of millions of HIGH-FREQUENCY, razor-thin SEDIMENTARY LAYERS. The texture mimics "eroded sandstone strata" or "glacial erosion patterns", creating an incredibly dense, rhythmic, striated look. NOT round noodles, NOT tubes. These are SHARP, thin, stacked sheets of material. Material: Translucent white jade (Subsurface Scattering) with a matte finish. It looks hard but lets light penetrate softly. Deep gorges hold pooled gradients of cyan and mineral green. The scale is massive, emphasized by microscopic pine trees on the ridge edges. Soft volumetric lighting, ethereal atmosphere, 8k, Octane render, ultra-detailed texture.

Create a high-end food advertising poster in a dark, moody restaurant atmosphere featuring a dish called {DISH_NAME}. Automatically analyze the dish name and infer the most common and culturally accurate ingredients associated with it, then visually showcase those ingredients around the dish in a dynamic and appetizing way. Display the dish name in large, bold, modern typography at the top of the image. The dish should be ultra-realistic, freshly prepared, centered on a wooden table, photographed from a slightly low angle with cinematic lighting, warm highlights, soft rim light, deep shadows, and shallow depth of field. Use a dark, blurred background with a subtle spotlight on the food. Style should match premium restaurant or fast-food advertising, hyper-realistic food photography, rich colors, glossy textures, professional studio quality. Aspect ratio 1:1 or 4:5, no watermark, no unrelated text.

{ "prompt": "ultra-realistic close-up food photography of four gourmet donuts displayed together on a rustic wooden board, each with a unique style and topping: (1) classic glazed donut with shiny golden glaze; (2) chocolate-frosted donut with dark chocolate drizzle and crushed nuts; (3) strawberry cream-filled donut with powdered sugar and fresh strawberry slices on top; (4) matcha green tea donut with white chocolate drizzle and pistachio crumbs, slight steam rising, glossy textures on frosting, spoon lifting a bite from one donut revealing soft fluffy interior, warm bakery environment with soft background blur featuring coffee cups and baking ingredients, golden hour lighting, cinematic composition, photorealistic, ultra-detailed textures, 8k resolution, shot on Sony A1 85mm f/1.4, razor-sharp focus on all donuts, rich warm color palette with chocolate browns, pastel pinks, greens, and golden tones", "negative_prompt": "blurry, lowres, jpeg artifacts, cartoon, illustration, drawing, painting, plastic looking, fake food, uneven shapes, bad proportions, text, watermark, signature, logo, over-saturated, washed out, harsh flash, cold lighting, 3d render, cgi, toy food, miniature food", "aspect_ratio": "4:5", "seed": -1, "cfg_scale": 7.5, "steps": 40, "sampler": "DPM++ 2M Karras or Euler a", "clip_skip": 2 }

A poetic cinematic portrait of a young woman standing alone in a vast snowy landscape, surrounded by distant frost-covered mountains. Her long auburn hair is caught mid-motion by strong winter wind, strands flowing naturally across her face. She wears a dark wool coat, minimal makeup, pale freckled skin, and a quiet, introspective expression. Soft snowfall drifts through the air. Muted natural color palette with cold greys, soft whites, and warm copper hair tones. Shallow depth of field, natural diffused daylight, overcast sky, fine film grain, subtle motion blur in hair and snow, cinematic realism, emotional storytelling, editorial photography, 85mm lens look, ultra-detailed, moody atmosphere, high realism, 8k quality,

a sleek rocket ship soaring through deep space, every surface lavishly encrusted with faceted gemstones that catch the light, shimmering with a holographic sheen and cascading sparkles, distant nebulae and starfields glowing in the background, dramatic high-contrast lighting, ultra-detailed, vibrant and fantastical sci-fi scene

Do this for Death Star: <instruction> 1. The Semantic Inference Engine (The Masterpiece): Input A is a Famous Building, City, or Fictional Structure (e.g., The Colosseum, Hogwarts, The Death Star). Analyze the subject to determine the Model Aesthetic : The Material: Identify the primary modeling medium. (e.g., Ancient Ruins -> Plaster/Clay. Modern City -> White Foamcore/Acrylic. Fantasy -> Wood/Moss). The Blueprints: Identify the reference documents on the table. (e.g., Vintage Parchment Maps vs. Blue Cyanotypes vs. Digital CAD printouts). 2. The Workshop : Goal: "Cinematic Artisan" Portrait. The Room: A cluttered, high-ceilinged Model Maker's Studio. Wooden shelves line the walls, filled with "Previous Projects" (other models in the same genre). The Character: an input inspired maker character wearing a relevant apron and an appropriate Visor or protective tool The Table: A scarred wooden workbench covered in scale modeling tools: X-Acto knives, glue bottles, sanding blocks, and sawdust. 3. The Narrative : The Pose: The craftsman is leaning in close to the "Model" (Input A), which sits centrally on the table. The Interaction: He is delicately placing a Tiny Detail Component onto the structure using tweezers or his fingers. (e.g., Planting a miniature tree, placing a tiny flag, or gluing a microscopic figure). The Scale: The model looks grand but miniature; the human hands look massive and weathered in comparison. 4. Lighting & Atmosphere: Lighting: Volumetric God Rays (Tyndall Effect). A strong, singular shaft of warm sunlight streams in from a high window (top left/center), illuminating the floating dust particles and spotlighting the model. Contrast: The workshop corners are in deep shadow; the work surface is bright. Palette: Warm, dusty Earth Tones (Browns, Golds, Beiges). Output: ONE image, 16:9 Aspect Ratio, Cinematic Photography, "National Geographic" aesthetic, Volumetric Lighting. </instruction>

Input Variable: [INSERT PODCAST NAME] System Instruction: Act as a Surrealist Set Designer. Create a 2x2 Grid. Curation: Identify 4 "Hidden Gem" podcasts similar to the Input. Media Inference: For each podcast, determine the physical medium that best represents its recording style (e.g., Vinyl, Magnetic Tape, Fiber Optic Cable, Wax Cylinder). The Composition: The recording medium is Unspooling or Unraveling into the air. The Mutation: As the medium unspools, it physically morphs into the Setting of the podcast. Texture Logic: The environment retains the texture of the medium (e.g., A glossy black "Tape" river; A glowing "Fiber Optic" forest; A carved "Vinyl" canyon). The Interaction: Tiny figures navigate this twisting, floating ribbon of media. Label: Written on the spool/connector: [Podcast Name]. Output: 2x2 panel, each panel for a gem podcast

Create a serene, top-down portrait of a young man lying peacefully in a lush meadow filled with wildflowers. He rests on his back among tall green grass, white daisies, and scattered pink blossoms, completely immersed in nature. One arm is bent behind his head in a relaxed pose, while the other gently holds a small bouquet of freshly picked daisies against his chest. He has soft, natural facial features, light stubble, and slightly tousled dark hair. His eyes are closed with a faint, content smile, conveying calmness, tenderness, and quiet joy. His expression feels intimate and emotionally warm, as if caught in a private moment of rest and reflection. He wears a textured mauve or dusty-rose knit sweater paired with clean white trousers, creating a gentle contrast against the vibrant greens and florals. The outfit feels cozy, romantic, and timeless. The scene is bathed in warm, natural daylight, soft afternoon, sunlight filtering through leaves and petals. Highlights gently kiss the skin and flowers, while shadows remain soft and organic. Foreground leaves and flowers partially frame the subject, adding depth and a dreamy, cinematic feel. Colors are rich yet natural: lush greens, creamy whites, soft yellows, and muted pinks, with a harmonious, film-like color grade.