GPT Image 1.5 prompts

A highly structured JSON prompt for GPT Image 1.5 to generate an ultra-photorealistic editorial image of a Scandinavian woman in a specific 'accidental fall' pose, focusing on natural skin texture, clinical lighting, and minimal studio env…

An extremely detailed, structured prompt for an ultra-realistic cinematic selfie. It specifies complex elements like pose, detailed facial features (preserving identity), outfit, environment (rainy night city with a police chase), high-con…

A prompt instructing the AI to create an infographic showing and labeling the internal and external components of a typical 5G smartphone, used here to test the models' ability to handle technical diagrams and labeling.

A prompt for generating an image of a cross-stitch design featuring a Christmas elf in anime style working at a guitar store, complete with a festive border.

A Japanese prompt instructing the AI to create an image of a character where lines point to different parts of the body, and the features of each part are explained in Japanese text.

A prompt used to generate a character sheet in both Nano Banana and GPT Image 1.5, based on an existing character image. The prompt requests a 5-section sheet with a full front shot and four medium close-up shots showing different angles (…

A user on Reddit asked ChatGPT to generate an anime character version of themselves using the ImageGen 1.5 model. This is a common use case for character transformation prompts.

A straightforward cinematic prompt for generating a moody image of a woman walking through a rainy street at night, focusing on specific visual elements like neon reflections, raindrops, soft bokeh, and a blue-red color contrast.

A prompt designed for GPT Image 1.5 to extract and separate individual clothing items from a person in the original image and display them laid out on a bed, showcasing the model's object extraction capabilities.

A simple prompt used with GPT Image 1.0 (and tested against 1.5) to convert an existing image into a night scene, specifically requesting warm lighting.

A highly specific prompt for generating a 2x2 grid of monumental desert ruins, where each ruin is constructed in the style of Cairo monumental stone and represents a major mathematical symbol (Integral, Sigma, Pi, Penrose Triangle).

A prompt designed to achieve a less 'AI-generated' look by specifying the image style as a photograph taken by the original iPhone, featuring a dog on a skateboard.

A Japanese prompt used with GPT Image 1.5 to generate a dynamic photo of a person performing a basketball dunk, focusing on capturing the sense of motion and action.

A prompt used with GPT Image 1.5 to create a 4K wallpaper featuring bright, vibrant colors suitable for a festive, celebratory mood. The user also mentions using an Image Extender tool to resize the output.

A highly detailed, structured prompt designed to transform an existing character (Aniko) into a maximalist, high-energy Harajuku Decora and Y2K fashion icon poster art. It emphasizes visual information overload, dynamic composition, and sp…

A highly detailed, structured prompt comparing GPT Image 1.5 and Gemini Nano Banana Pro. The prompt aims to generate a hyper-realistic, cinematic portrait of a rugged man in a worn leather jacket, set in a moody outdoor mountain landscape…

A detailed image generation prompt designed to create a hyper-realistic photo of a white woman's hand adorned with elegant diamond bracelets, focusing on luxury jewelry editorial lighting and a high-end fashion photography aesthetic agains…

A set of conversational instructions used with GPT Image 1.5 to correct details in a generated image, specifically adjusting the position of a ring, the color of a scarf, and extending a sleeve.

A highly specific prompt for generating a moody, cinematic portrait of a young male model. It details the lighting (harsh directional light from the left), pose (clutching collar, tense jawline), and a complex motion blur effect slicing ho…

A detailed prompt for generating a cinematic, high-contrast portrait of a young male model in a black suit, featuring harsh directional lighting, a tense expression, and a horizontal motion blur effect slicing across the face.

A complex instructional prompt designed for image editing, requiring multiple specific text replacements, a date change, a visual element replacement (screen content), and a subject pose change (boy facing viewer with thumbs up). This demo…

A prompt designed to test the advertising capabilities of GPT Image 1.5 by generating a vertical format (9:16) ad for Oreo cookies featuring Santa Claus. This was used in a head-to-head comparison with Nano Banana Pro.

A simple image editing prompt for GPT Image 1.5, instructing the model to change the outfit of the subject in an uploaded image to a tuxedo and a bow tie. This highlights the model's ability to handle basic inpainting and style changes.

A prompt designed for GPT-Image 1.5 to create a 3x3 grid of portraits based on an uploaded photo. It requires maintaining the subject's exact likeness while generating nine different facial expressions, presented with a clean, modern aesth…

A creative prompt instructing the model to convert an uploaded image of a couple into an ultra-realistic portrait rendered as a blue ink spill on paper, maintaining the faces but depicting a romantic pose and specific traditional clothing.

A prompt designed to test the technical instruction following of image models by requesting a 3x2 matrix showing six different views (front, back, left, right, top, bottom) of a highly detailed robot character, emphasizing PBR quality.

A Japanese prompt instructing GPT Image 1.5 to generate a cover for TIME magazine's '100 People of the Year' issue, specifying that all titles and text should be in English for a professional look.

A prompt designed for product photography, generating an ultra-high-resolution image of a reference object fully submerged in crystal-clear turquoise water, focusing on intricate caustic light patterns and ethereal elegance.

A highly detailed, cinematic prompt for GPT Image 1.5, requesting a photorealistic, documentary-style rendering of a 12th-century Javanese battlefield at dawn, focusing on a Pendekar Jawa (Javanese warrior) in silhouette, emphasizing histo…

A detailed prompt for generating a portrait in the style of 1960s comic book pop art. It specifies using bold black contour lines, flat saturated primary colors, dense halftone shading, expressive facial features, and a full-bleed, graphic…

This prompt is used with ChatGPT Image to generate an image where every detail is aligned and every color tells a story, suggesting a highly detailed and aesthetically pleasing visual output.

A conceptual prompt strategy for GPT-Image 1.5 focusing on achieving artistic output by controlling information density. Step 1 involves increasing information (dynamic angles, complex poses) and Step 2 involves reducing information (omitt…

A highly detailed prompt for generating a candid, cinematic lifestyle photograph. It specifies the subjects (tired father and young child), setting (dimly lit living room, vintage couch), actions (father holding remote, child eating chips)…

A prompt aimed at changing a character's visual style to a 'more realistic visual' while keeping the original atmosphere of the anime or manga intact. This is useful for adapting characters to different media styles.

A simple prompt used to generate an image of a Christmas wreath suitable for hanging on a front door, showcasing a basic image generation request.

A complex, multi-step prompt designed to generate a cohesive 3x3 cinematic contact sheet from a single uploaded source image. It instructs the model to internalize the subject, scene, and emotional state, and then create a set of images ca…

A highly detailed and complex prompt designed to create a seamless 16:9 composite image divided into four distinct quadrants, each representing a different brand (Luckin, Google, Coke, McDonald's) using intricate paper-cut aesthetics and s…

A simple Japanese prompt requesting a cool illustration of the Ouroboros—two snakes eating each other—symbolizing a closed system.

A user tested GPT Image-1.5 by asking it to create a detailed advertisement for a Christmas outfit, noting that the model successfully generated the ad but missed the price tag, which is crucial for a complete advertisement.

A prompt used to generate an 'educational AI art' collage consisting of 10 elements, where the subject is specified as the tenth piece stacked in the center. The user notes this is a custom prompt based on a previous reference.

A descriptive prompt for generating a horror-themed image of a dimly lit, empty hallway lined with antique portraits, culminating in an intimidating, unknown creature with luminescent red eyes.

A Japanese prompt describing two characters: an exorcist with fluffy rabbit ears and a rabbit familiar who is also an apprentice and wants to take a bath. This prompt was used to test GPT Image 1.5's ability to handle complex character des…

A creative prompt template, referred to as the 'FCG method,' which instructs the AI to generate a set of images depicting a crazy story unfolding over specific time intervals: 5 minutes, 10 minutes, 1 hour, and 3 hours.

A prompt used with GPT-4o to generate an image depicting a girl on a rooftop looking up at the sky, capturing a sense of longing and liberation.

This text describes the desired output style for an image generation task using GPT Image 1.5, focusing on retaining the decadent mood and texture of 90s cinema while reconstructing the figures realistically. This description acts as a hig…

This entry indicates that a specific image was created using a photo as a reference combined with a Grok Imagine prompt. The prompt itself is not provided, but the structure implies an image-to-image or visual reference workflow.

A highly detailed, structured prompt designed to transform an existing character (Aniko) into a maximalist, high-energy Harajuku Decora and Y2K fashion icon poster art. It emphasizes visual information overload, dynamic composition, and sp…

A highly detailed prompt for generating a hyperrealistic dark fantasy portrait of a mysterious hooded mage, focusing on intricate details like runic tattoos, wet fabric textures, glowing magical effects, and dramatic volumetric lighting in…