CHEER LIN / ARCHIVE

GPT Image 2.5 Prompt Library

315 image generation and editing entries across storyboards, sprites, products, portraits, and typography. Published prompts retain their original wording; other entries display source notes

GPT Image 2.5Image Generation & EditingPrompts315 cases

Category guide

All prompts (315)

Waterfall portrait in a white shirt dress
Title

Waterfall portrait in a white shirt dress

Prompt
Hyper-realistic IMAX-level Netflix-style cinematic moody waterfall portrait, 9:16 vertical. Use the uploaded image as the primary visual reference and preserve the exact facial identity, proportions, and defining features. Create a woman wearing a long white shirt-style dress with a simple collarless neckline, loose full-length sleeves and lightweight fabric that becomes soaked and clings naturally in the rushing water. Her entire body is positioned in a clear side profile facing left, seated directly beneath a powerful waterfall, with her back close to the dark rock behind her. Her torso remains upright with a slight backward lean, while her hips rest on a low rock hidden beneath the foamy water. Both elbows are bent and lifted toward her face, with both hands gently placed against the sides of her head near her ears, fingers spread through her wet hair. Her head tilts slightly backward and toward the left, her eyes closed, cheeks raised and mouth open in a spontaneous, joyful laugh. Her legs are bent and crossed in front of her: one lower leg descends almost vertically into the rushing water, while the other crosses diagonally over it, its foot extending forward toward the right. The lower legs and feet are intermittently obscured by splashing foam. Her long, dark hair is completely wet and swept backward by the falling water, with loose strands framing her face and trailing over her shoulders. Fair luminous porcelain skin with a bright ivory to light beige tone and a neutral-cool undertone, retaining realistic wet skin texture and tiny droplets across her face, arms and legs. The woman sits fully underneath the waterfall, with a heavy curtain of water pouring over her head and shoulders and bursting outward around her body; turbulent white foam fills the foreground while dark, rugged rock walls frame the cascade. Cool diffused daylight, icy silver highlights on the falling water and deep slate-blue shadows across the rocks create a dramatic atmosphere. Apply moody cinematic colour grading with cool blue-grey tones, crisp white water, subdued saturation and naturally warm skin tones, preserving fine details in the wet shirt, hair and airborne droplets. Negative prompt: front-facing body, standing pose, incorrect crossed legs, wrong hand placement, short top, distorted hands, bad anatomy, artificial water, plastic skin, text, watermark.
Tags
GPT Image 2.5

Public image prompt from the source, preserved in its original language.

Needs your own references The prompt refers to image, video or audio inputs that are not included on this site

Accidental phone-drop snapshot
Title

Accidental phone-drop snapshot

Prompt
【主体】 × iPhone 意外录像截帧 × 发现手机后的自然拾取动作 × 贴地超低机位广角透视 9:16竖版,手机意外掉落在户外地面,后置摄像头朝上持续录像。成年女性发现手机后准备拾取,视线落向手机,身体重心随动作变化,动作尚未完成。 天空、树冠与城市建筑随机进入背景,保留偶然取景。 iPhone 原生视频随机截帧,避免摆拍姿势、完美对称构图、过度锐化与人为颗粒。
Tags
GPT Image 2.5

Public image prompt from the source, preserved in its original language.

Title

POPEYE & OLIVE OYL character bible sheet

Prompt
Create a premium cinematic character bible sheet for POPEYE & OLIVE OYL. Use uploaded character sheets as strict visual reference for both characters. Do not change either's appearance. LAYOUT: Single unified wide format. Both characters large and dominant filling most of the sheet — close up from mid thigh up, taking up almost the entire frame. Popeye on the left, Olive Oyl on the right. Deep navy and volcanic red watercolor explosion radiating behind both. TOP LEFT: Large bold brushstroke text POPEYE in deep navy. Below small text: HOT SAUCE / PREHISTORIC JUNGLE. TOP RIGHT: Large bold brushstroke text OLIVE OYL in volcanic red. Below small text: THE RESCUE / THE JUNGLE. CHARACTERS: Both cropped close from mid thigh up filling most of the sheet. Popeye on the left — enormous forearms, face still slightly red, fresh toothpick in corner of mouth, steam fading from ears, one arm flexing, one arm around Olive Oyl. Olive Oyl on the right — black hair bun, red top, black skirt with yellow trim, one enormously long arm thrown dramatically around Popeye, wide relieved expression. Behind both — faint silhouettes of dinosaurs and volcanic mountains fading into the watercolor. BOTTOM CENTER: Bold text POPEYE & OLIVE OYL in volcanic red slightly worn. Tagline: HE CHUGGED THE HOT SAUCE. THE DINOSAURS DIDN'T STAND A CHANCE. COLOR PALETTE: Deep navy, volcanic red, white. OVERALL: Clean white background, deep navy and volcanic red watercolor explosion radiating behind both characters, faint dinosaur silhouettes in background, bold flat color blocking, chunky simplified forms, hard edge shadows, thick black outlines, cinematic cel-shaded 3D anime, hand-painted textures, not cartoon not Disney not Pixar, print ready.
Tags
GPT Image 2.5

Prompt from the original author’s reply; both the main post and reply are linked.

Needs your own references The prompt refers to image, video or audio inputs that are not included on this site

@TechieBySA
Landmark behind a giant zipper
1 / 2
Title

Landmark behind a giant zipper

Prompt
Create a breathtaking, photorealistic surrealist photograph of a giant metallic zipper partially opening the surface of an ordinary real-world environment, revealing an actual place from [COUNTRY / CITY / LANDMARK] behind it A giant, physically realistic silver zipper runs diagonally across the scene. Its metallic teeth separate two edges of the outer surface, creating a large, natural-looking opening. The zipper must look like a real physical object, with brushed metal texture, fine scratches, accurate reflections, and realistic shadows. INSIDE THE OPENING: Reveal a breathtaking, photorealistic view of [COUNTRY / CITY / LANDMARK] as it actually exists in the real world. If the subject is a country, show an authentic landscape featuring its real geographical features, traditional architecture, recognizable landmarks, and genuine cultural details. If the subject is a city, show its actual streets, real buildings, recognizable skyline, authentic transportation, and characteristic urban atmosphere. If the subject is a landmark, reproduce its real architecture with accurate proportions, materials, colors, windows, structural details, and surrounding environment. Every building, street, vehicle, and architectural feature must look believable and consistent with the real location. Use actual-world architectural references rather than inventing structures. THE VISUAL MAGIC: The only surreal element is the giant zipper opening reality. The location inside must look like an authentic photograph taken by a professional travel photographer. The real landscape extends naturally behind the zipper opening, with realistic perspective, atmospheric depth, natural sunlight, and physically accurate shadows. The zipper edges cast believable shadows onto the revealed scene, making the illusion convincing. PHOTOGRAPHIC STYLE: Ultra-photorealistic travel photography, natural daylight, realistic atmospheric perspective, true-to-life colors, physically accurate textures, professional full-frame camera quality, sharp architectural details, natural contrast, realistic depth of field, and premium editorial photography. COMPOSITION: Vertical 4:5 composition. The giant zipper occupies the foreground and opens wide enough to reveal the location clearly. The real-world destination is the main subject, immediately recognizable and visually dominant. Clean composition with no unnecessary decorative elements. STRICT REALISM RULES: No fantasy architecture or imaginary buildings. No cartoon, illustration, painting, or CGI-looking landscapes. No exaggerated colors, artificial glowing lights, or magical particles. No distorted buildings, melted structures, or impossible roads. No invented landmarks or inaccurate architectural details. No extra zippers or unrelated objects. No text, labels, borders, or watermarks. Final result: A visually extraordinary yet convincing photorealistic image in which a giant zipper reveals a real place hidden behind reality. The concept is surreal; the destination itself is completely realistic.
Tags
GPT Image 2.5

Public image prompt from the source, preserved in its original language.

Food photography and ingredient journal poster
1 / 4
Title

Food photography and ingredient journal poster

Prompt
Create a premium 3:4 vertical editorial food-art poster for [COUNTRY / CITY / LOCAL DISH / FOOD SUBJECT]. TEXT-TO-IMAGE ONLY. Generate the entire composition from the written dish or food subject alone. Do not request, upload, use, or depend on any reference image or existing photograph. Automatically identify the dish’s authentic appearance, ingredients, textures, and cultural identity. CORE CONCEPT: Create one elegant travel-food editorial poster combining realistic food photography in the upper section with a handmade ingredient-observation sketchbook in the lower section. Feature ONE recognizable plate or bowl containing a single dish, with at least three visually distinct ingredients. UPPER SECTION — REALISTIC FOOD PHOTOGRAPHY: Occupy approximately 48% of the canvas height. Create a beautifully composed, authentic photographic view of the specified local dish, preserving believable ingredient textures, natural lighting, realistic proportions, and culturally accurate presentation. Use subtle food-magazine color grading and extremely light film grain. Keep the entire plate or bowl visible. SECTION DIVIDER: Separate both sections with a straight, clean horizontal boundary. A very thin warm-ivory paper strip is optional. No gradients, blending, or elements crossing the divider. LOWER SECTION — INGREDIENT OBSERVATION SHEET: Use warm ivory drawing paper with visible fibers, fine grain, faint smudges, and subtle handmade irregularities. Create a sparse, asymmetrical food-field-note composition rather than repeating the full photographic dish. Place ONE small, simplified but recognizable illustration of the complete dish near the center. Keep it secondary and compact, occupying approximately one-third of the lower section’s width. Around it, illustrate 3–6 individually selected ingredients inferred from the dish. Use small watercolor specimens, cut-surface studies, herb sketches, grain clusters, sauce swatches, or distinctive ingredient shapes. Arrange them at varied sizes and angles with generous breathing room. DRAWING STYLE: Matte gouache, delicate black-ink outlines, slightly hesitant hand-drawn strokes, dry-brush gaps, uneven pigment coverage, soft watercolor edges, visible paper texture, and restrained natural colors. Keep ingredients recognizable without excessive detail. Add a few fine arrows and small handwritten numbers connecting specimens to the miniature dish. TYPOGRAPHY: Use minimal, understated editorial annotations only when helpful. Short ingredient names or tiny handwritten notes are acceptable. Keep all text small and secondary; avoid unnecessary titles or decorative lettering. COLOR PALETTE: Derive the colors naturally from the dish, balanced by warm cream, muted earth tones, and subtle ink details. FINAL AESTHETIC: Authentic local food photography × handmade ingredient studies × refined travel-food editorial × collectible culinary field notebook. AVOID: Multiple dishes, duplicate plates, mixed meals, inaccurate ingredients, excessive garnish, crowded layouts, gradients, bleeding transitions, glossy digital painting, flat vector infographics, 3D rendering, commercial menu styling, excessive text, logos, watermarks, and interface elements. FORMAT: 3:4 vertical portrait, approximately 48% photographic food scene above and 52% illustrated ingredient-observation sheet below, crisp horizontal separation, generous negative space, premium publication quality, ultra-high definition.
Tags
GPT Image 2.5

Public image prompt from the source, preserved in its original language.

Geometric character illustration poster
1 / 3
Title

Geometric character illustration poster

Prompt
Create a finished editorial character poster inspired by the uploaded reference image. Preserve the recognizable identity, hairstyle, clothing details, pose, and important visual characteristics of the subject while translating them into a sophisticated flat geometric illustration. FORMAT LOCK Vertical 3:4 composition. Edge-to-edge warm off-white paper background. One standalone poster. Clean editorial layout. CHARACTER One adult character constructed from a few oversized asymmetric geometric shapes. Slightly exaggerated proportions, small head, elongated capsule-like limbs, simple mitten-style hands, and one oversized object integrated naturally into the pose and silhouette. Preserve the subject's recognizable appearance while simplifying details into graphic shapes. STYLE Flat geometric illustration, asymmetric proportions, crisp color blocks, soft airbrushed shading, subtle dark-to-color spray gradients where shapes overlap, fine print grain, delicate stippling mostly inside darker areas, refined contemporary editorial poster aesthetic. FACE & EXPRESSION Minimal black cut-paper facial features: simple crescent eyes and minimal mouth. The emotion should be communicated primarily through the character's full-body gesture and silhouette rather than detailed facial expression. PALETTE Warm off-white #F3F1EB, near-black 10100F, hot pink #F553D2, warm red #F42726, cobalt 293EA1, emerald 13785D. Keep colors flat and graphic with controlled soft gradients only where forms overlap. Keep the character and oversized object as the main visual focus. Use generous negative space and a balanced editorial arrangement. The object should visibly affect the overall silhouette and feel integrated into the character's pose. NEGATIVE PROMPT No extra text, logos, outlined cartoon style, glossy 3D, realistic skin, photorealism, coarse canvas texture, busy background, extra limbs, distorted anatomy, excessive facial detail, photorealistic shading, random objects, clutter, watermark.
Tags
GPT Image 2.5

Public image prompt from the source, preserved in its original language.

Needs your own references The prompt refers to image, video or audio inputs that are not included on this site

Title

MR BEAN & CONOR MCGREGOR character bible sheet

Prompt
Create a premium cinematic character bible sheet for MR BEAN & CONOR MCGREGOR. Use uploaded character sheets as strict visual reference for both characters. Do not change either's appearance. LAYOUT: Split screen VS format. Two halves divided by a bold dramatic dividing element in the center. LEFT SIDE — MR BEAN: Bold dramatic bright red watercolor splash radiating behind him filling the entire left side. Large bold brushstroke text MR BEAN top left in bright red. Below small text: THE LEGEND / LONDON. One massive dramatic cropped hero image of Mr Bean from mid-thigh up — brown tweed jacket, bright red tie, giant inflatable hammer in one hand, small stuffed toy under the other arm, completely calm and unbothered expression. CENTER: Bold dramatic VS in emerald green. Below it small text: LONDON STREET / MIDDAY / NO RULES. RIGHT SIDE — MCGREGOR: Bold dramatic emerald green watercolor splash radiating behind him filling the entire right side. Large bold brushstroke text MCGREGOR top right in emerald green. Below small text: THE NOTORIOUS / DUBLIN. One massive dramatic cropped hero image of McGregor from mid-thigh up — green UFC fight shorts, black MMA gloves, heavy chest tattoos, fists raised in fighting stance, intense focused expression, dripping confidence. BOTTOM CENTER: Bold text THE BEAN VS THE NOTORIOUS in emerald green. Tagline: ONE HAS NEVER LOST A FIGHT. ONE HAS NEVER BEEN IN ONE. SAME RESULT. OVERALL: Clean white background, bold dramatic bright red watercolor explosion left side, bold dramatic emerald green watercolor explosion right side, dramatic green center divider, bold flat color blocking, chunky simplified forms, hard edge shadows, thick black outlines, cinematic cel-shaded 3D anime, hand-painted textures, not cartoon not Disney not Pixar, print ready.
Tags
GPT Image 2.5

Prompt from the original author’s reply; both the main post and reply are linked.

Needs your own references The prompt refers to image, video or audio inputs that are not included on this site

@TechieBySA
Landscape photo to travel-art poster
Title

Landscape photo to travel-art poster

Prompt
Create a sophisticated contemporary travel-art poster from the uploaded londscape photograph. Strict 3:4 vertical format. The upper half must remain the original landscape photogroph. Preserve all important environmental structures, natural textures, lighting, shadows, horizon, architecture, people and perspective. Do not regenerate the scene. The lower half reconstructs the landscape's strongest narrative element using delicate photographic fragments, halftone printing, subtle paper fibers and fading edges. Use a single horizontal geometric color field sampled from the landscape. Let the reconstructed subject or horizon physically intersect this field. Keep the background most white with muted tones derived from the original image. Avoid saturated colors. The result should resemble a limited-edition museum travel poster rather than a conventional photo collage.
Tags
GPT Image 2.5

Public image prompt from the source, preserved in its original language.

Needs your own references The prompt refers to image, video or audio inputs that are not included on this site

Night-flash ponytail portrait
1 / 4
Title

Night-flash ponytail portrait

Prompt
Ultra-photorealistic candid night portrait of an adult woman with fair porcelain glass skin, luminous dewy complexion, soft natural rosy blush, softly defined brows, subtle brown eyeshadow, natural lashes, and peachy-nude glossy lips. Extremely long, thick espresso brown hair in a high messy ponytail with loose curls strands blowing across her face. Wearing a delicate white spaghetti-strap dress, backless, thick gold hoop earrings, subtle piercings and gold rings. She gently holds a huge bouquet of white daisies with yellow centers close to her face, partially covering her mouth and lower face, gazing softly toward the camera with a dreamy candid expression. Dark outdoor nighttime background with warm distant bokeh lights, direct on-camera flash, slight motion blur, subtle film grain, intimate 35mm analog-camera aesthetic, cinematic low-light atmosphere, ultra-detailed. Add a small handwriting watermark written dev at the bottom left. Ar 3:4.
Tags
GPT Image 2.5

Public image prompt from the source, preserved in its original language.

Watermelon snack editorial illustration
1 / 2
Title

Watermelon snack editorial illustration

Prompt
Create an original vertical 9:16 semi-realistic hand-drawn editorial illustration of a quirky young adult woman caught in a funny, spontaneous moment while eating a slice of watermelon. She has realistic human facial anatomy, naturally flushed cheeks, detailed expressive hazel eyes, a few freckles, and messy dark auburn hair loosely tied in a high ponytail with a mustard-yellow ribbon. She wears a vivid plum-purple blouse with loose cream-colored stripes. Her body leans forward over a kitchen table, her head tilted slightly to one side, eyes glancing sideways with a comically suspicious expression, cheeks slightly puffed, and lips pressed together as she tries not to laugh. One hand holds a juicy watermelon slice close to her mouth while the other rests casually on the table beside a small ceramic plate. The image must look like a traditional artist's expressive sketchbook illustration, with loose charcoal and graphite strokes, rough ink outlines, visible construction marks, scratchy cross-hatching, textured watercolor washes and opaque gouache brushwork. Realistic skin shading and believable human hands, painterly facial details, imperfect sketch lines extending beyond the figure, lively spontaneous brush marks. Background: a loosely sketched muted dusty-lavender kitchen wall with pale mint and warm ivory brush strokes, faint illegible handwritten-style artistic scribbles, and unfinished pencil marks. Use a fresh, vibrant color palette of plum purple, watermelon red, mustard yellow, dusty lavender, pale mint, and warm peach skin tones. Medium close-up composition, face large and sharply readable, natural human proportions, expressive eyes and mouth, visible paper texture, sophisticated hand-illustrated magazine art aesthetic. Keep the character visually dominant and the background loose and sketchy. IMPORTANT STYLE: semi-realistic human illustration, NOT cartoon, NOT anime, NOT 3D, NOT glossy digital art, NOT a photograph. Preserve the raw, loose, textured, imperfect hand-drawn quality of traditional editorial sketch art. Original character, original pose, original outfit, original composition.
Tags
GPT Image 2.5

Public image prompt from the source, preserved in its original language.

Autumn watercolor travel journal
1 / 4
Title

Autumn watercolor travel journal

Prompt
Create a premium 4:5 vertical autumn travel-journal artwork for [COUNTRY / CITY / LOCATION / SUBJECT]. TEXT-TO-IMAGE ONLY — generate every element from the written destination or subject alone. Do not request, upload, use, or depend on any reference image, existing photograph, or source image. MASTER STYLE LOCK: Hand-painted autumn watercolor on warm ivory paper, minimalist travel-editorial illustration, transparent watercolor washes mixed with soft gouache, delicate imperfect ink outlines, visible paper grain, natural pigment blooms, soft edges, muted autumn colors, generous negative space, warm nostalgic atmosphere, premium handcrafted art-book aesthetic. COMPOSITION: Divide the vertical canvas into two equal halves using a precise 50/50 left-right split. LEFT HALF — REALISTIC DESTINATION: Generate an original, natural-looking photograph-style scene of [COUNTRY / CITY / LOCATION / SUBJECT]. Automatically determine its recognizable architecture, scenery, atmosphere, and natural lighting. Preserve realistic textures, authentic colors, and believable perspective. Keep this entire half clean, with no text, labels, illustrations, borders, or decorative elements. RIGHT HALF — AUTUMN WATERCOLOR JOURNAL: Reimagine the same destination as a delicate, handmade autumn watercolor illustration. Use simplified rounded architecture, warm glowing windows, a small tree with dusty rose, apricot, and amber foliage, a few drifting leaves, and subtle watercolor clouds. Include only a few meaningful local details, keeping the composition airy and uncluttered. COLOR PALETTE: Caramel brown, dusty rose, apricot orange, creamy yellow, muted maple red, warm ivory, and faint lavender-grey shadows. Use soft, low-saturation afternoon light. No neon colors or harsh black. HANDMADE DETAILS: Add one or two narrow kraft-brown or dusty-pink washi tape strips, a small pressed autumn leaf, and a few delicate hand-drawn stars or arrows. Integrate these details naturally without overcrowding the illustration. DESTINATION LABEL: Place a small cream-colored paper label near the upper-left corner of the RIGHT HALF. Main title: [LOCATION NAME] Subtitle: [ENGLISH LOCATION NAME] Use rounded, handwritten Japanese-inspired lettering for the main title and smaller, widely spaced handwritten English lettering for the subtitle. Keep spelling accurate and typography legible. SIGNATURE LABEL: Near the bottom-right corner of the RIGHT HALF, add a small kraft-paper label with exactly this text: RYOBIYORI No. [NUMBER] / [YEAR] Use subtle, fine dark-brown lettering. STRICT LAYOUT RULES: All text, labels, washi tape, and decorative elements must remain entirely within the right half. The left half must contain only the original generated realistic destination scene. Never reproduce or modify an existing photograph. PAPER & FINISH: Warm ivory watercolor-paper texture, delicate fibers, subtle grain, organic pigment edges, and a carefully assembled travel-journal feeling. FINAL AESTHETIC: Realistic destination imagery × autumn watercolor illustration × handmade travel journal × nostalgic seasonal palette × premium editorial design. AVOID: Reference-image requests, image-editing instructions, existing-photo preservation, unequal panel widths, text crossing the center divider, excessive decoration, oversaturated colors, glossy digital painting, 3D rendering, misspelled text, extra logos, watermarks, and interface elements. OUTPUT: 4:5 vertical portrait, exact 50/50 left-right composition, realistic generated destination scene on the left, watercolor travel-journal illustration on the right, ultra-high definition.
Tags
GPT Image 2.5

Public image prompt from the source, preserved in its original language.

Old house as a miniature model
1 / 2
Title

Old house as a miniature model

Prompt
参考上传的老房子照片,以1:24比例精准复刻为博物馆级超写实实体微缩模型,严格还原建筑结构、屋顶、门窗、墙体及原始环境。精细呈现斑驳墙皮、风化砖瓦、裂缝、青苔、灰尘等岁月痕迹,材质真实自然,具有极致手工制作质感。模型置于精致的矩形沙盘底座上,45度俯视三分之四视角,完整展示建筑全貌。专业微距摄影,柔和自然光,真实阴影,中性灰摄影棚背景,8K超高清。强调真实实体模型效果,避免塑料玩具感、卡通感及3D渲染感。
Tags
GPT Image 2.5

Public image prompt from the source, preserved in its original language.

Needs your own references The prompt refers to image, video or audio inputs that are not included on this site

Showing 12 / 315