灵感图库
探索和使用精选的 AI 图像生成模板 (共 6,519 个)
Nano Banana Pro 4-Panel Comic Explanation
A prompt instructing an AI to create a four-panel comic strip explaining the 'Nano Banana Pro' product, utilizing a specified character for the illustration.
Inverted Cityscape Car Chase Prompt for Nano Banana Pro
A highly detailed, structured JSON prompt designed for the Nano Banana Pro model to generate a cinematic still frame. The scene features a bright orange, low-profile sports car accelerating rapidly along a spiral highway above a vertically mirrored, inverted cityscape reminiscent of Blade Runner 2049. The prompt specifies detailed composition, lighting, camera settings, and color grading for a dynamic, high-contrast, realistic result.
Photorealistic Watermelon Fruit Carving Prompt
A detailed image generation prompt designed to create a photorealistic image of a whole watermelon carved in the intricate Thai fruit carving style. The carving features a specific character description, surrounded by elegant floral and geometric reliefs, emphasizing the contrast between the green skin, white rind, and red flesh. The final image is set on a silver tray surrounded by fresh fruits.
Image Repair and Inpainting Prompt
A simple instruction prompt for Nano Banana Pro to naturally repair a damaged part of an image, specifically a broken finger, demonstrating the model's inpainting capabilities.
Realistic Interior to Transparent Miniature Diorama Prompt
This is the second part of the two-step prompt sequence, using the previously generated realistic interior image as a reference. The goal is to transform that realistic room into a highly detailed, transparent miniature architectural model or diorama, similar to a museum exhibit. The prompt emphasizes preserving the exact layout and proportions while focusing on high-end industrial design aesthetics, transparent acrylic/glass casing, realistic miniature materials, and sophisticated internal/external lighting contrast.
Structured Prompt for Automated Manga Generation
A structured prompt template used in a pipeline combining Claude and Nano Banana Pro to automatically generate cooking manga. It defines parameters for the scene, characters, dialogue, and specific prompts for both the manga panel and the real-life cooking image.
Chibi Character 'Tehepero' Pose Generation Prompt
A detailed structured prompt for Nano Banana Pro to generate a chibi character performing the 'Tehepero' pose (winking and sticking out the tongue slightly). It specifies facial expressions, body pose, camera angle (high angle, 24mm equivalent), and composition (full body, centered, simple white background). This prompt requires an existing character image as input.
Chalkboard Math Problem Image Generation
A prompt for Nano Banana Pro to generate an image of a math problem written and solved on a green chalkboard with chalk, aiming for a realistic, imperfect texture (chalk dust, pressure marks) to enhance engagement.
Manga-style Sketch Prompt for Nano Banana Pro
A simple text prompt designed for Nano Banana Pro to generate images in a loose, comforting, and sketch-like style, similar to a professional manga artist's casual drawing. The example shows it correctly interpreting the age of a character (Makipon) and providing a soft drink instead of alcohol, highlighting the model's contextual awareness.
Detailed Calendar Design Prompt for Nano Banana Pro
An extremely detailed, structured prompt (likely YAML or JSON format) for Nano Banana Pro to generate a sophisticated, commercial-quality design calendar for February 2026. It specifies layout, color palette, typography rules, composition, and data mapping for dates and holidays, aiming for a premium modern minimal aesthetic.
Chibi Character Tehepero Pose Prompt for Nano Banana Pro
A detailed prompt for Nano Banana Pro to generate a super-deformed (chibi) character performing the 'Tehepero' pose (sticking out the tongue and winking). The prompt specifies the character's action, gaze, camera angle, perspective, lens, and composition, emphasizing a big head and small body proportion, and requires an attached image for the character reference.
Low-Angle Full-Body Portrait on Concrete Steps
A prompt for generating a high-resolution, low-angle, full-body shot of a woman sitting on outdoor concrete steps, specifying her attire (oversized flannel, dark trousers, chunky sneakers) and the moody lighting against an apartment building backdrop.
Premium Top-Down Product Marketing Shot Prompt
A prompt template for Nano Banana Pro designed to generate high-end editorial product photography shots from a top-down view. It allows customization of the product and surface texture, emphasizing 3D realism, luxury aesthetic, and natural sunlight.
Geospatial and historical image generation prompt
A simple Spanish prompt for Nano Banana to create an image based on specific geographical coordinates (Jerusalem), date (April 3, 33 AD), and time (3:00 PM), focusing on historical and location specificity.
Intricate Paper Folding Morphological Transformation Prompt
This is a detailed, multi-section JSON-like prompt structure for Nano Banana Pro, designed for complex image transformations. It specifies an anchor image, intricate paper folding morphology, material physics (cardboard/washi paper), specific illumination (hard shadows, architectural lighting), and a render stack (tilt-shift, macro, stop-motion aesthetic). It also includes a negative prompt to exclude unwanted elements.
Exploded Burger Infographic and Product Shot
A dual-image prompt for generating a high-quality product photograph of a zinger burger and a hyper-realistic exploded vertical infographic composition of a gourmet burger. The prompt also includes a specific motion instruction for video generation, focusing on controlled, precise separation of layers.
Raw Reality TV Broadcast Screenshots (2x2 Grid)
A complex, structured prompt designed to generate a 2x2 grid of images mimicking raw, unpolished TV broadcast footage, specifically in the style of a reality show like Big Brother. It mandates on-screen graphics (timestamps, camera IDs) and specific visual aesthetics for four different scenes: Domestic Friction, Confessional Booth, Night Vision Secrets, and The Explosion.
Structured Prompt for Ultra-Photorealistic Candid Street Photography
A highly structured JSON prompt template for generating ultra-photorealistic, candid street photography. It specifies detailed output settings (aspect ratio, resolution, film grain), camera physics (35mm lens, eye-level), and environment details (urban clutter, natural light), aiming for a documentary, emotionally true feel.
3x3 Cinematic Fashion Collage on a Rooftop at Sunset
A detailed prompt for generating a 3x3 grid cinematic fashion collage featuring a young man on a rooftop at sunset. It specifies the subject's streetwear, golden hour lighting, and nine distinct poses/actions across the grid, ensuring consistency in identity, environment, and lighting across all frames.
Cinematic Lifestyle Portrait in a Luxury Car
A highly detailed, structured prompt for generating an ultra-realistic cinematic lifestyle portrait of a woman (Taylor Swift is mentioned as an example) in the back seat of a luxury car at night. It mandates 1:1 identity preservation, realistic skin texture (visible pores), specific makeup, and a pose holding a single pale pink rose, emphasizing a seductive, editorial mood.
Pop Art Plush Product Shot of a Beverage Can
A prompt for generating a product photograph of a beverage can rendered entirely as a vibrant, fluffy plush fabric sculpture. It specifies a bold Pop Art and Memphis design style, clean studio lighting to emphasize texture, and high color saturation for a modern commercial aesthetic.
Food Photography and Video Prompt for Blueberry Bagel and Iced Coffee
A multi-modal prompt combining instructions for image generation (lifestyle shot of a blueberry bagel being dusted with sugar), video generation (gentle stirring of iced coffee with controlled fluid motion), and audio generation (sophisticated cafe jazz). It focuses on candid, lifestyle documentary aesthetics and realistic micro-movements.
Woven Wooden Twig Portrait with Identity Lock
A prompt for generating a hyper-realistic portrait where the face and hair are constructed entirely from small, woven wooden twigs and dry bark texture. The key instruction is to use an uploaded photo for 100% identity lock, ensuring the generated wooden sculpture is clearly recognizable as the reference person.
Food History Diorama in Mandala Format (Spicy Chicken Wings)
A complex, multi-layered prompt for generating a hyper-realistic 'Food History Diorama' arranged in a Mandala format, using spicy chicken wings as the subject. It requires five concentric rings, each representing a different historical era (Modern, Commercial, Domestic, Industrial, Ancient), populated by tiny 1:87 scale chef figures dressed appropriately for that era, with specific instructions for lighting, scale, and texture changes across the rings.