Prompts

GPT Image 2.5 API awesome prompts

🚀 Curated GPT Image 2.5 3000+ prompt library — pixel-perfect text rendering, cross-image consistency, and commercial-grade illustration. Free & open source.

Results: 3701
Profile / AvatarModel: gpt-image-2

Floating Stone Cube Challenge

Generates a vertical anime-style social media image of a massive cracked stone cube floating above ruins with dramatic Japanese challenge text.

Floating Stone Cube Challenge
Prompt

Create a vertical smartphone screenshot-style anime fantasy scene with black app UI bars at the top and bottom. In the main image, show a dramatic low-angle view looking up through a ruined canyon of towering cracked stone walls under a vivid deep blue sky with bright white clouds and sunlight flaring from the upper right. A massive cubical stone block floats in the center near the top, made of many cracked rectangular slabs, with sharp rim lighting, chipped edges, and small debris fragments breaking off. Count 6 floating stone objects total: 1 enormous central cube, 1 medium rectangular block below-left of center, 1 small cuboid at upper right, 1 small cuboid near lower center, 1 small cube near bottom center, and 1 tiny distant block among the debris. Add dynamic dust, sparks, and motion streaks to make the block feel suspended or falling. Place bold vertical Japanese manga text with thick black fill, white outline, and warm glow: on the right side, {argument name="right vertical text" default="どうだい?"}; on the left side, {argument name="left vertical text" default="壊してみないかい?"}. The art style should be cinematic high-detail anime, saturated colors, extreme perspective, dramatic lighting, crisp stone texture, and a sense of awe and challenge. Include a minimal social-media viewer overlay: black top bar with a white X icon at top left and the number {argument name="top counter" default="38"} at top right, plus a black bottom bar with small like and dislike icons in a rounded dark pill and Android navigation symbols.

Author:太郎
Social Media PostModel: gpt-image-2

Aesthetic Doodle Portrait

A creative prompt that blends ultra-realistic photography with playful hand-drawn doodles and typography, perfect for a modern Gen Z social media aesthetic.

Aesthetic Doodle Portrait
Prompt

Low-angle aesthetic portrait of {argument name="subject" default="subject from reference image"}, standing under a bright blue sky filled with soft, fluffy clouds. The subject is wearing a pastel bucket hat, oversized gray sweatshirt, and round sunglasses, styled in minimal fashion. Relaxed and calm expression, natural pose. Above and surrounding the subject are playful white doodle illustrations seamlessly integrated into the sky: cute cartoon characters, rocket, robot, skull, stars, paper airplane, sparkles, and whimsical abstract shapes. Handwritten typography floating in the sky with the text “{argument name="first name" default="custom name 1"}” and “{argument name="second name" default="custom name 2"}”, drawn in soft white ink, bubbly, dreamy, and slightly imperfect hand-drawn style. Clean composition, airy spacing, pastel tones, soft natural lighting, minimal shadows, crisp details. Kawaii doodle overlay combined with ultra-realistic photography style. Instagram-style edit with subtle film grain and soft glow. Modern Gen Z aesthetic, light blue dominant palette, dreamy and playful mood, high dynamic range, editorial photography, 4K resolution. Aspect ratio 3:4

Author:Ciri
Profile / AvatarModel: gpt-image-2

Synthetic iPhone Photos Metadata Screenshot

Creates a realistic rebuilt iPhone Photos details screen from a reference screenshot, preserving metadata, map UI, and controls while subtly regenerating the image content.

Synthetic iPhone Photos Metadata Screenshot
Prompt

Using the provided iPhone Photos screenshot as the reference, recreate it as a cleaner, high-fidelity synthetic screenshot while preserving the same overall content: the cloud photo, the Photos app metadata panel, the map, date/time, filename, camera details, and bottom controls. Subtly rebuild the cloud image rather than copying it exactly, with a slightly wider-feeling crop, sharper modern rendering, and a more polished dark iOS interface. Keep the visible text and metadata unchanged, including {argument name="date and time" default="Sunday • Sep 3, 2017 • 9:13 PM"}, {argument name="filename" default="IMG_2139"}, {argument name="device name" default="Apple iPhone 7 Plus"}, {argument name="location" default="Del Valle"}, and {argument name="photo specs" default="Telephoto Camera — 56 mm ƒ2.8; 12 MP • 4032 × 3024 • 1.6 MB; ISO 20; 57 mm; 0 ev; ƒ2.8; 1/618 s"}. Include exactly 5 bottom toolbar icons: share, heart, information, adjustment sliders, and trash. Keep the map section with the centered photo-pin thumbnail and the visible road shields 71 Texas, 183 toll, 45 toll, and 130 toll.

Author:8fstudioz
Product MarketingModel: gpt-image-2

Watercolor Anime Café Colorization

Transforms a grayscale café line drawing into a warm, richly colored hand-painted anime illustration while preserving the original composition.

Watercolor Anime Café Colorization
Prompt

Using REFERENCE_0 as the exact line-art base, colorize the illustration into a richly detailed hand-painted watercolor/gouache anime illustration. Preserve the original composition, perspective, character pose, café setting, furniture, bookshelves, window, plants, food, and all line details, but replace the grayscale pencil look with warm natural colors and painterly texture. Use a cozy vintage café palette: honey-brown wood, cream and gold kimono fabric, blonde hair, blue and multicolored books, green foliage, yellow flowers, warm pendant lights, coffee and pastries. Keep the masked face area as a flat plain rectangle with no facial features. Add depth through soft sunlight from the window, gentle highlights on glass and porcelain, layered shadows, visible brush strokes, and high-detail color accents while maintaining the delicate original linework.

Profile / AvatarModel: gpt-image-2

Doodled Strawberry Matcha Drink Photo

Generates a realistic handheld cafe drink photo enhanced with whimsical hand-drawn doodles for social media food visuals.

Doodled Strawberry Matcha Drink Photo
Prompt

Create a realistic shallow-depth-of-field street photo of a hand holding a clear plastic cup of iced layered drink in front of a small outdoor beverage stall. The main drink is a tall transparent cup with a domed flat lid, filled with three distinct layers: deep red berry syrup at the bottom, creamy white milk with red streaks in the middle, and chunky bright green matcha on top; the cup is held by a left hand in a muted gray-green sleeve. Keep the background softly blurred: an outdoor counter with syrup bottles on the right, a white cloth banner reading large playful black text "Eh!" and "Chocolat", plus small menu words "Chocolate", "Maring", and "Refreshing", a pink menu board behind it, and a white illustrated drink menu board on the left. Add playful hand-drawn doodles directly interacting with the drink and scene, as if drawn over the photo: one smiling strawberry character riding a red-and-white candy-striped ribbon swirl around the cup, one smiling green leaf character riding a small red-and-white rocket launching from the cup toward the upper right, three visible candy-striped ribbon arcs wrapping around or streaming from the drink, one dotted multicolor path curling down the left side of the hand, one small handwritten "Phew!" with three blue sweat drops near the thumb, one arch of small cookie or chocolate-chip doodles above the lid with a tiny square chocolate character at the top, four yellow star doodles around the banner, and three small doodled drink icons on the banner matching the labels: a brown chocolate drink, a yellow maring drink, and a green refreshing drink. Use cheerful black marker outlines, red, green, yellow, and blue accent colors, whimsical motion lines, and integrate the doodles so they mimic the drink's flowing layers and energetic fresh flavor. The scene should feel like a candid food-and-drink social media photo enhanced with cute animated doodles, preserving natural lighting, realistic hand anatomy, and the original composition centered on the cup. Customize the drink flavor as {argument name="drink flavor" default="strawberry matcha milk"}, the main banner text as {argument name="banner text" default="Eh! Chocolat"}, the handwritten exclamation as {argument name="exclamation text" default="Phew!"}, the doodle mascot pair as {argument name="doodle mascots" default="smiling strawberry and green leaf"}, and the stall setting as {argument name="stall setting" default="outdoor beverage stall"}.

Author:Oogie
Product MarketingModel: gpt-image-2

Cute Superhero Drink Doodle Photo

Generates a realistic handheld café drink photo enhanced with playful white doodles that turn the cup into a whimsical superhero character.

Cute Superhero Drink Doodle Photo
Prompt

Create a realistic handheld street-food photo of a layered iced drink in a clear plastic takeaway cup with a domed flat lid, held close to the camera by a left hand wearing a muted gray-green long sleeve. The drink has three vivid layers: a deep ruby-red fruit base at the bottom, a cloudy white creamy middle with red streaks, and a thick bright matcha-green top with small dark specks. Set the scene outdoors in front of a small dessert drink stall, with shallow depth of field and natural daylight; the background should be softly blurred but still show a white sign reading “Eh! Chocolat,” a pink menu board, syrup bottles lined up on the counter, a green bottle at the right, and a colorful menu display at the left. Add playful white hand-drawn doodles directly over the photo so the cup becomes a cute superhero character: one kawaii face on the cup with two eyes, one open smile, two blush-mark groups, and one outlined star on its body; two noodle-like arms, two simple legs, and two small boots; one flowing cape extending to the right; one wand or magic staff with a circular tip; two small hearts, one four-point sparkle, and nine short motion/emphasis strokes around the cup. Keep the doodles casual, imperfect, bright white, and integrated with the subject as if sketched on top of the photograph. Use a vibrant candid smartphone photography look, high contrast, crisp focus on the drink and hand, blurred café-stall background, and a cheerful whimsical mood. Customize the stall sign text as {argument name="stall sign text" default="Eh! Chocolat"}, the drink top color as {argument name="drink top color" default="bright matcha green"}, the drink bottom color as {argument name="drink bottom color" default="deep ruby red"}, the doodle color as {argument name="doodle color" default="white"}, and the sleeve color as {argument name="sleeve color" default="muted gray-green"}.

Author:Oogie
Profile / AvatarModel: gpt-image-2

Pixar Style 3D Man Avatar

A simple prompt for creating 3D character avatars in a Pixar-like aesthetic, suitable for profile pictures or interface icons.

Pixar Style 3D Man Avatar
Prompt

3D avatar of a {argument name="subject" default="man"}, with a {argument name="expression" default="happy face"} on a white background, no necks. Conceptual playlist style digital art with {argument name="style" default="Pixar quality aesthetics"}

Author:aiiStudio
Social Media PostModel: gpt-image-2

Miniature Girl on Giant Football

A whimsical fantasy prompt featuring a miniature girl on a giant football, emphasizing scale contrast and soft lighting.

Miniature Girl on Giant Football
Prompt

A cute {argument name="character description" default="miniature blonde girl with long wavy blonde hair"} sitting comfortably on a {argument name="object" default="giant football"} held between large human fingers, exaggerated playful scale contrast, dreamy surreal scene, glossy textures, vibrant colors, soft glowing sky, aesthetic Instagram-style composition, whimsical fantasy realism, hyper-detailed outfit with a {argument name="outfit" default="red top and denim shorts"}, smooth skin, digital art style, slightly exaggerated cute proportions, sharp focus with soft dreamy background.

Author:Amy G
Profile / AvatarModel: gpt-image-2

Crouching Tribal Warrior Portrait

A cinematic realistic prompt for generating a dramatic feathered tribal warrior crouching on a mountain ridge with stormy atmosphere.

Crouching Tribal Warrior Portrait
Prompt

Create a cinematic, hyper-realistic portrait of a {argument name="subject" default="powerful indigenous tribal warrior"} crouching on a grassy mountain ridge, facing the camera in a low, dominant pose. The subject has dark weathered skin with white ritual body-paint markings across the arms, chest, shoulders, and legs, and wears a dramatic feather headdress made of many tall dark green, brown, and gray feathers radiating upward. The face is deliberately hidden by a centered, flat dark-brown square censor block. The warrior holds one long wooden spear vertically in the left hand, with a dark stone spearhead wrapped in rough cord. Clothing and accessories are primitive and handmade: layered fur and bark-fiber cloak over the shoulders, rope armbands and wrist wraps, a rough woven loincloth, dangling cords, and a necklace of carved bones, teeth, and beads. The figure is muscular, grounded, and tense, with one hand planted on the mossy earth in the foreground. Set the scene against distant misty green mountains under a dramatic {argument name="sky mood" default="stormy gray overcast sky"}. Use moody natural light, high contrast, desaturated earthy colors, sharp texture detail on skin, feathers, rope, fur, mud, grass, and stone, shallow depth of field, epic documentary photography style, vertical portrait composition, realistic 85mm lens look, no text.

Author:ARION
Product MarketingModel: gpt-image-2

Hand-Drawn Doodles Overlay

A creative prompt for adding expressive, hand-drawn doodles to an existing photo while preserving the original subject's identity, pose, and lighting for a playful, sketchbook-like aesthetic.

Hand-Drawn Doodles Overlay
Prompt

Transform the uploaded image by keeping the {argument name="subject" default="subject"} exactly as they are—preserve their identity, pose, composition, and lighting without any distortion or alteration. Build on top of the original photo by adding expressive, hand-drawn doodles that interact playfully with the subject. Let the illustrations respond to the scene, trace or exaggerate gestures, extend movement, highlight contours, or introduce imaginative elements that feel connected to what’s happening in the frame. The doodles should feel intentional, as if they were sketched directly onto the image in a spontaneous yet thoughtful way. Use a loose, imperfect drawing style with slightly uneven strokes, organic lines, and a casual, sketchbook-like vibe. Incorporate {argument name="notes" default="handwritten notes or captions"} around the image that match the mood or narrative—keep them witty, context-aware, and naturally playful rather than generic. Make sure the added elements complement the photo instead of overpowering it. Aim for a visually balanced composition with a clean, high-resolution finish and colors that blend harmoniously with the original scene.

Author:Jack
Product MarketingModel: gpt-image-2

Childlike Crayon Drawing Re-creation

A prompt to transform an existing image into a simplified crayon drawing, as if created by a young child with playful elements.

Childlike Crayon Drawing Re-creation
Prompt

Please recreate the entire image in a {argument name="style" default="crayon drawing style"}. Simplify the details so that it looks like it was drawn by a {argument name="age" default="10-year-old child"}. Do not use the original colors from the image. Make the overall look feel like it was drawn on a sheet of white paper, with a very cute and playful vibe. You can add {argument name="decorative elements" default="adorable elements such as flowers, candies, stars, clouds, etc."} to give it a childlike and innocent feel. 4:5

Product MarketingModel: gpt-image-2

High-End Fashion 'Stride' Poster

A high-fashion advertisement prompt featuring giant oversized props and bold typography for a modern editorial look.

High-End Fashion 'Stride' Poster
Prompt

A high-end fashion advertisement poster. {argument name="model" default="A young woman in an all-white oversized sweatsuit"} leans casually against a giant white sneaker that is 3x her height, the sneaker has "{argument name="text" default="STRIDE"}" written on its side. Light blue studio background with reflective glossy floor. Bold massive white sans-serif typography "{argument name="text" default="STRIDE"}" fills the entire background behind her. Tagline at bottom center in white italic font: "{argument name="tagline" default="Walk into your next chapter."}" Small text top-right corner reads "Designed with GPT Image 2" in grey. Photorealistic, editorial magazine style, medium format camera look.

Author:𝐌
Comic / StoryboardModel: gpt-image-2

Action Storyboard Simplification

A workflow prompt designed to simplify a complex reference sheet into specific action shots showing dynamic martial arts impacts and thrilling cinematic movement.

Action Storyboard Simplification
Prompt

Take this reference sheet and simplify it to {argument name="shot count" default="8 shots"} and show actual {argument name="action" default="impacts for each blow"}. Show me a thrilling cinematic scene with dynamic martial arts action and impacts following.

Profile / AvatarModel: gpt-image-2

Ultra Rare Imagination God TCG Card

Generates a holographic Japanese-style ultra rare trading card featuring a cute anime deity who creates floating worlds.

Ultra Rare Imagination God TCG Card
Prompt

{"type":"ultra rare Japanese trading card game character card","style":"premium anime fantasy TCG card, glossy holographic rainbow foil, ornate black and gold frame, highly detailed illustration, dramatic magical lighting, crisp printed typography, collectible card layout","canvas":"vertical 2:3 card with rounded corners and thick metallic gold border","character":{"name":"{argument name=\"character name\" default=\"想像神ちゃん\"}","reading":"そううぞうしん","role":"cute childlike imagination deity","appearance":"small anime girl floating in space, pale skin, short bob haircut with blonde-to-ash-blue gradient, blue eyes, calm slightly unimpressed expression, two flower hair clips on the right side, one red flower and one pale daisy","outfit":"oversized white hoodie with colorful patchwork sleeves, geometric multicolor buttons and pixel motifs, red shorts or skirt, white socks with blue bands, red-and-green sneakers","pose":"seated weightless pose, legs extended forward, both hands open as if creating worlds"},"main_art":{"background":"cosmic blue and violet vortex with sparkling stars, aurora rainbow light, circular portal behind the character","world_objects":{"count":8,"items":["top-left translucent cube world with rainbow city texture","left-middle floating castle island above the character hand","upper-right green-blue floating cube fragment","right red crystal cube","right-middle large green village island","lower-left castle island partly behind vertical text","lower-center ocean cube with a white city","bottom-center green fantasy island"]},"mood":"godlike but adorable, world-creation magic, overwhelming power"},"card_layout":{"top_left":{"rarity":"UUR","star_count":5,"description":"large rainbow-gradient UUR letters with five rainbow stars beneath"},"top_center":{"title":"{argument name=\"card title\" default=\"想像神ちゃん\"}","subtitle":"~世界を想像して創造する神様~"},"top_right":{"class_icon":"神","attribute_label":"属性:想像"},"left_vertical_quote":"{argument name=\"vertical quote\" default=\"想像は、世界を創るいちばんの魔法。\"}","stats_panel":{"position":"lower right","level":"LV. 99/99","count":5,"stats":[{"label":"想像力","value":"9999"},{"label":"創造力","value":"9999"},{"label":"発想力","value":"9999"},{"label":"共感力","value":"9999"},{"label":"無限力","value":"∞"}]},"ability_panels":{"position":"lower left","count":2,"abilities":[{"header":"必殺技","name":"イマジネーション・ジェネシス","description":"カード全体に無限の可能性を広げる光を放ち、すべての想像を現実に変える。"},{"header":"神の力","name":"インフィニット・ワールド","description":"このカードが場にある限り、プレイヤーは毎ターン新しい世界を創造できる。"}]},"profile_panel":{"position":"lower right below stats","title":"プロフィール","text":"世界のあらゆる「はじまり」を想像することで生み出した神様。退屈が大嫌いで、いつも新しいアイデアを探している。『一緒に、なにか想像してみる?』"},"footer":{"count":4,"items":["種族:神","性別:なし","イラスト:想像神の世界","IG-0001-UUR"]}},"rendering_notes":"Make the card look like a real high-rarity Japanese mobile game trading card, with gold filigree corners, black enamel UI boxes, white Japanese text, rainbow foil reflections, tiny sparkles, and ultra-sharp anime line art. Do not include any external watermark."}

Profile / AvatarModel: gpt-image-2

Cozy Cinematic Home Workspace Portrait

A warm and inviting portrait prompt of a young woman in a home office with soft golden ambient lighting and bokeh effects.

Cozy Cinematic Home Workspace Portrait
Prompt

Cinematic indoor portrait of a young woman sitting on a chair and looking back over her shoulder with a soft, natural smile, cozy home workspace environment, warm ambient lighting with golden tones, blurred computer monitor in the background, soft bokeh lights, relaxed atmosphere. Natural skin texture, minimal makeup, casual outfit, shallow depth of field, 85mm lens, soft focus, warm color grading, storytelling composition, ultra-realistic, 4K, lifestyle photography style. Moody cinematic portrait of a young woman leaning on a desk with her chin resting on her hand, gentle smile, warm dim lighting from desk lamps creating soft glow, blurred computer screen with editing interface in background, cozy nighttime workspace, intimate and relaxed mood. Soft shadows, creamy bokeh, natural skin tones, slightly messy hair, casual black outfit, shallow depth of field, 85mm lens, film-like color grading, ultra-realistic, 4K, aesthetic lifestyle photography

Author:Taaruk
Profile / AvatarModel: gpt-image-2

Shynloc Visual Identity Board

Generates a polished brand identity guideline page for an AI-driven folded N logo with logo concept, diagram, color palette, and construction details.

Shynloc Visual Identity Board
Prompt

{"type":"clean corporate visual identity system presentation board","style":"premium minimalist brand guideline page, white background, thin light-gray divider lines, precise grid layout, editorial typography, black and vivid orange accents, subtle shadows, InDesign-style brand manual, high resolution","brand":{"name":"{argument name=\"brand name\" default=\"Shynloc\"}","system title":"{argument name=\"system title\" default=\"SHYNLOC VISUAL IDENTITY SYSTEM\"}","page label":"PAGE 01","theme":"AI-driven folded N logo concept"},"layout":{"canvas":"portrait A4 brand guideline sheet with generous margins","top bar":{"count":3,"items":["left header text: SHYNLOC VISUAL IDENTITY SYSTEM","thin horizontal rule across the page","right page marker: PAGE 01 with 01 in orange"]},"hero section":{"position":"upper half","title":"BRAND IDENTITY","subtitle":"Logo Concept","right tagline":"AI-DRIVEN FOLDED N with a small orange slash mark","main logo":"large centered primary wordmark reading {argument name=\"primary logo text\" default=\"ShyNloc\"}; black italic rounded script letters, with the central N replaced by a sharp origami folded orange symbol; the folded N uses angular facets, gradients from deep orange to golden yellow, and crisp triangular planes"},"sections":[{"title":"PRIMARY LOGO","position":"center divider label above lower content","count":1,"labels":["large primary ShyNloc wordmark"]},{"title":"SYMBOL MARK","position":"lower left large panel","count":1,"labels":["standalone folded orange N mark with subtle gray shadow"]},{"title":"CONCEPT DIAGRAM","position":"lower right upper panel","count":3,"labels":["gray letter N labeled LETTER N IDENTITY","gray AI labeled ARTIFICIAL INTELLIGENCE","orange folded N symbol labeled FOLDED N AI-DRIVEN"],"diagram text":"N + AI = folded N symbol"},{"title":"CONCEPT","position":"beneath concept diagram","count":1,"labels":["paragraph explaining the logo concept"]},{"title":"COLOR PALETTE","position":"bottom left panel","count":2,"labels":["Onyx Black #0A0A0A","Shynloc Orange gradient #FF7A00 – #FFB400"],"swatches":"one black rectangle and one orange gradient rectangle divided into warm orange bands"},{"title":"CONSTRUCTION","position":"bottom right panel","count":1,"labels":["geometric construction drawing of the ShyNloc logo"],"details":"gray wordmark overlaid with orange construction lines on the folded N, dotted guides, bounding boxes, diagonal guide lines, and two 45° angle labels"}],"footer":{"count":4,"items":["small black square containing orange folded N icon","text block: SHYNLOC VISUAL IDENTITY SYSTEM","text block: VERSION 1.0 / MAY 2024","text block: DESIGNED FOR THE FUTURE. DRIVEN BY INTELLIGENCE. with orange slash at far right"]}},"typography":{"headings":"uppercase geometric sans serif, wide tracking","body":"small condensed sans serif, precise corporate guideline copy","wordmark":"custom bold italic rounded script for the black letters"},"colors":{"background":"warm white","primary black":"{argument name=\"black color hex\" default=\"#0A0A0A\"}","primary orange":"{argument name=\"orange color hex\" default=\"#FF7A00\"}","gold highlight":"#FFB400","divider gray":"#D8D8D8","diagram gray":"#B8B8B8"},"concept paragraph":"The Shynloc logo centers on a bold folded N that intelligently integrates the essence of AI. The lower diagonal fold forms an A, while the vertical stroke completes the I, creating a forward-looking mark that communicates innovation, intelligence, and precision. Origami-inspired geometry reflects adaptability, engineering, and future-ready thinking.","rendering notes":"make all text crisp and legible, align everything to a strict grid, use subtle orange gradient facets on the folded N, keep the page uncluttered and luxurious, no mockup perspective, flat front-facing presentation board"}

Author:ShyNloc
Page 4 of 232