PROMPT LIBRARY

Image prompts for GPT Image 2.5 and Nano Banana 2.1.

Each prompt is quoted word for word from the model maker's guide or from a creator's post, next to the output it produced. Click a picture to watch or view the original.

Maintained on GitHub: awesome-image-video-prompts awesome-prompts

Nine failed photos of any character

Nine failed photos of any character

GPT Image 2.5 · Text to image · 2,646likes · EN

Seven words. The author's point: fewer restrictions give more authentic-looking mistakes.

Source: @DeepBlueX0 on X
Upper half photo, lower half watercolour travel sketch

Upper half photo, lower half watercolour travel sketch

GPT Image 2.5 · Edit your photo · 2,264likes · EN

A vertical split where the photo dissolves into watercolour. Works in Gemini, Grok and ChatGPT, per the author.

Source: @MahnoorAi12 on X
Minimal lifestyle art poster from a real photo

Minimal lifestyle art poster from a real photo

GPT Image 2.5 · Edit your photo · 1,698likes · EN

A fill-in template: swap `[PERSON / SUBJECT / MOMENT]`. Keeps the subject recognisable while dropping the photographic background.

Source: @Naiknelofar788 on X
Sixteen panel dance pose sheet of anime girl

Sixteen panel dance pose sheet of anime girl

GPT Image 2.5 · Edit your photo · 1,270likes · EN

Locks identity to one uploaded reference, then scripts sixteen numbered moves from opening pose to finishing silhouette so the grid reads as one routine. Fixed camera, white background and an explicit avoid list keep scale, costume and panel layout consistent.

Source: @Mayz1169 on X
Highland cow papercraft cube toy template

Highland cow papercraft cube toy template

GPT Image 2.5 · Edit your photo · 1,178likes · EN

The prompt fixes the source photo as the only reference for the animal and lists each flat piece with folds, tabs and cut edges. A closing list of things not to change keeps the animal's identity and fur texture intact.

Source: @icreatelife on X
Turn each photo into a quiet editorial illustration poster

Turn each photo into a quiet editorial illustration poster

GPT Image 2.5 · Edit your photo · 1,165likes · EN

Opens by forbidding collages, one poster per photo, and closes with a row of mood words the result should feel like.

Source: @icreatelife on X
Camera roll of anyone's day, from one image

Camera roll of anyone's day, from one image

GPT Image 2.5 · Edit your photo · 762likes · EN

Same idea with stricter rules: the upload is the only subject, and only minimal iOS chrome may appear.

Source: @Mayz1169 on X
Six-photo portrait grid of one person

Six-photo portrait grid of one person

GPT Image 2.5 · Text to image · 651likes · EN

Describes the person once in detail, then six shots. Asks for no gutters so the grid reads as one seamless image.

Source: @john_my07 on X
Pencil handwritten essay on lined paper with coffee stain

Pencil handwritten essay on lined paper with coffee stain

GPT Image 2.5 · Text to image · 614likes · EN

The prompt fixes the medium (pencil on 8.5x11 lined paper) and the capture method (phone photo). It asks for stroke variance and one small flaw, a coffee stain in a named corner, which gives the realism.

Source: @chetaslua on X
Photo to minimalist paper-textured art cover

Photo to minimalist paper-textured art cover

GPT Image 2.5 · Edit your photo · 561likes · EN

Organised under headings (composition, palette, texture). Ends by naming the reference point: an independent art-book cover.

Source: @Ciri_ai on X
Building photo to halftone travel-collage poster

Building photo to halftone travel-collage poster

GPT Image 2.5 · Edit your photo · 497likes · EN

Long spec with a closing priority rule: architectural accuracy first, the halftone treatment second.

Source: @Naiknelofar788 on X
Full-body fashion illustration in textured geometric style

Full-body fashion illustration in textured geometric style

Nano Banana 2.1 · Text to image · 494likes · EN

Outfit described item by item, then a short ban list that rules out photorealism and 3D.

Source: @ChillaiKalan__ on X
Lakeside Photo Paired With Paper Illustration

Lakeside Photo Paired With Paper Illustration

GPT Image 2.5 · Edit your photo · 417likes · EN

A strict 50/50 split keeps the top as a faithful photo and turns the bottom into a small centered handmade illustration of the same scene. Capping the palette at 4 colors extracted from the photo, plus generous paper negative space, ties the two halves together.

Source: @harboriis on X
Sunlit woodland clearing with dense green foliage

Sunlit woodland clearing with dense green foliage

GPT Image 2.5 · Text to image · 415likes · EN

A single plain sentence that names the subject, the photo medium, and a highly detailed cue works as a noise artifact test, because dense repeating foliage exposes texture smearing and grain. The short wording leaves the model free to choose light and composition.

Source: @mark_k on X
Half photo, half black-ink illustration poster

Half photo, half black-ink illustration poster

Nano Banana 2.1 · Edit your photo · 414likes · EN

Asks for an exact 50/50 split, then specifies how each half treats the same subject.

Source: @ChillaiKalan__ on X
Your city as a tiny world growing out of a map

Your city as a tiny world growing out of a map

GPT Image 2.5 · Text to image · 413likes · EN

Swap `[COUNTRY]` and `[CITY]`. Optional landmark slots at the end let you steer what appears.

Source: @Preda2005 on X
Platinum bob model in cobalt circular opening

Platinum bob model in cobalt circular opening

Nano Banana 2.1 · Text to image · 370likes · EN

The prompt is split into labeled blocks for scene, outfit, pose, lighting, palette and photography style. Naming a lens, a medium-format look and a cool-versus-warm light split keeps the high-end editorial feel consistent.

Source: @Emmma___0 on X
Editorial poster of woman with roses and yellow bag

Editorial poster of woman with roses and yellow bag

GPT Image 2.5 · Edit your photo · 337likes · EN

The canvas is split into two exact halves. The top keeps the original photo untouched, and the bottom rebuilds the subject over one low saturation color window sampled from the photo. The subject crosses the window edge into white negative space, which gives depth and avoids a sticker look.

Source: @Ziha_ai on X
Photo to knitted stop-motion puppet bust

Photo to knitted stop-motion puppet bust

GPT Image 2.5 · Edit your photo · 327likes · EN

Very long material spec (yarn, stitches, felt) with a final rule not to invent anything the photo does not show.

Source: @oggii_0 on X
JSON prompt that reconstructs a photo shot for shot

JSON prompt that reconstructs a photo shot for shot

Nano Banana 2.1 · Edit your photo · 326likes · EN

Structured as JSON fields (framing, light, colour, negatives). The post compares the same JSON on Nano Banana 2.1 and GPT Image 2.5.

Source: @ViralOps_ on X
Low-angle dusk portrait with film grain

Low-angle dusk portrait with film grain

GPT Image 2.5 · Text to image · 318likes · EN

Under 600 characters. Camera angle, clothing detail, one light source, one film look.

Source: @chatgptpaglu on X
A full day as an iPhone Photos library screenshot

A full day as an iPhone Photos library screenshot

GPT Image 2.5 · Edit your photo · 309likes · EN

Recreates the real iOS Photos interface down to the status bar, filled with candid shots of the reference person.

Source: @BubbleBrain on X
Phone snapshot on a train platform

Phone snapshot on a train platform

GPT Image 2.5 · Text to image · 305likes · EN

Realism through phone-camera flaws: motion blur on the train, real mobile noise, slight wide-angle distortion.

Source: @MissDelulu9 on X
Woman in brown leather jacket at wine bar

Woman in brown leather jacket at wine bar

GPT Image 2.5 · Edit your photo · 298likes · EN

A reference photo is first analysed into a detailed JSON of color grading, lighting and exact colors. That JSON plus an attached product image then drives the generation, which avoids the flat AI look and keeps the style consistent.

Source: @Mho_23 on X
Photo turned into blue doodle couple

Photo turned into blue doodle couple

GPT Image 2.5 · Edit your photo · 280likes · EN

The prompt says what to preserve (people, pose, composition) and then simplifies the style with a tight palette of cobalt outlines, peach blush and golden hearts on cream paper. Negative constraints such as no gradients, no shading and no textures keep the result flat and hand drawn.

Source: @Ciri_ai on X
Photograph above, sketch below: an editorial template

Photograph above, sketch below: an editorial template

Nano Banana 2.1 · Text to image · 266likes · EN

Fill `[SUBJECT]` and `[SETTING]`. The ban list covers the usual split-poster failures: heavy borders, mismatched faces.

Source: @ChillaiKalan__ on X
Rusty red tram photo paired with paper illustration

Rusty red tram photo paired with paper illustration

GPT Image 2.5 · Edit your photo · 266likes · EN

The prompt splits a 3:4 canvas into two equal halves, with the original photo kept intact on top and a tiny four color paper illustration below. Capping the illustration at 10 to 20 percent of the bottom half and demanding warm off white negative space gives it an art book cover feel.

Source: @Ziha_ai on X
Second World: your photo, rebuilt as another world

Second World: your photo, rebuilt as another world

Nano Banana 2.1 · Edit your photo · 263likes · EN

A versioned JSON prompt (v1.2) written by @hx831126 and shared by @Just_sharon7. Includes its own self-check for seams and text.

Source: @Just_sharon7 on X
Half face, name stacked beside it

Half face, name stacked beside it

Nano Banana 2.1 · Edit your photo · 244likes · EN

Each letter of the name hides part of the other half of the face. Fill in name, backdrop and lighting.

Source: @Emmma___0 on X
Honey bee educational infographic with data modules

Honey bee educational infographic with data modules

GPT Image 2.5 · Text to image · 230likes · EN

A subject placeholder lets one template cover any species. It names every module (profile sidebar, facts list, size chart, rating, habitat icons, status slider) and fixes the palette to soft beige for a journal look.

Source: @iamsofiaijaz on X
Istanbul landmark sketch travel poster in red ink

Istanbul landmark sketch travel poster in red ink

GPT Image 2.5 · Text to image · 211likes · EN

A fill in the blank template swaps in the landmark, city and coordinates while fixed style terms keep the look consistent. Ivory paper, hatching, halftone, geometric shapes and a 4:5 ratio give the print feel.

Source: @Naiknelofar788 on X
Vintage film portrait framed by pink hibiscus flowers

Vintage film portrait framed by pink hibiscus flowers

Nano Banana 2.1 · Text to image · 206likes · EN

Shooting through foreground petals at eye level builds a natural floral frame, while hard rules like side part, one hand and head to chest crop keep the composition fixed. Film grain, halation and light leaks plus a long negative list steer the result away from a polished studio look.

Source: @ZaraIrahh on X
Shanghai skyline photo paired with isometric miniature diorama

Shanghai skyline photo paired with isometric miniature diorama

GPT Image 2.5 · Text to image · 200likes · EN

The poster is split into two stacked halves, a photoreal destination scene on top and an isometric diorama of the same scene on a dark slab below. Continuity rules, a bracketed location slot and fixed typography blocks with warm off white negative space keep the layout reusable for any city.

Source: @Goodmanprotocol on X
Fashion illustration in layered 3D paper-cut

Fashion illustration in layered 3D paper-cut

Nano Banana 2.1 · Text to image · 188likes · EN

Same structure as the geometric one, with the medium swapped. Good pair for comparing styles.

Source: @ChillaiKalan__ on X
Nostalgic magazine artbook spread from a photo

Nostalgic magazine artbook spread from a photo

Nano Banana 2.1 · Edit your photo · 183likes · EN

Keeps the original subjects and adds a sketched echo of them, laid out like a printed spread.

Source: @Just_sharon7 on X
Recreate any image as a newspaper collage

Recreate any image as a newspaper collage

Nano Banana 2.1 · Edit your photo · 171likes · EN

One sentence edit. Works on photos and illustrations alike.

Source: @aimikoda on X
Window Photo Continuing Into Ink Sketch Paper

Window Photo Continuing Into Ink Sketch Paper

GPT Image 2.5 · Edit your photo · 171likes · EN

A split canvas makes one structure, here the windowsill, cross the seam so the photo continues as paper and line drawing. A torn edge follows the source shape, and a short handwritten caption closes the scene.

Source: @harboriis on X
Cinematic chiaroscuro close-up of your photo

Cinematic chiaroscuro close-up of your photo

Nano Banana 2.1 · Edit your photo · 170likes · EN

One light from the side and an explicit request for visible pores to keep skin from looking airbrushed.

Source: @rovvmut_ on X
Over the shoulder flash selfie on boucle sofa

Over the shoulder flash selfie on boucle sofa

Nano Banana 2.1 · Edit your photo · 157likes · EN

The prompt splits the brief into labeled blocks for pose, hair, expression, outfit, background and camera, and a tight crop that cuts off half the face keeps the framing intimate. A long negative list plus identity preservation from the attached photo keeps the face, neck and hands believable.

Source: @Emmma___0 on X
Tree silhouette against huge sun over frozen lake

Tree silhouette against huge sun over frozen lake

GPT Image 2.5 · Text to image · 150likes · EN

The prompt is split into labeled blocks for sky, sun, tree, city, water, reflection and palette, so each element gets its own placement rules. A long negative list and explicit rules such as buildings never overlapping the tree or sun keep the symmetrical composition intact.

Source: @churvikv on X
Bakery logo on a transparent background

Bakery logo on a transparent background

GPT Image 2.5 · Text to image · EN

Asks for an original mark that reads at small and large sizes, then spells out the transparent background: clean alpha edges, no backdrop, no checkerboard. Pair it with background=transparent in the API. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Candid 35mm portrait of an old sailor mending a net

Candid 35mm portrait of an old sailor mending a net

GPT Image 2.5 · Text to image · EN

Skin texture, lens, light and grain are each named, and the last line rules out glamour and retouching, so the result reads as an unposed photo. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Cellular respiration diagram for a high school class

Cellular respiration diagram for a high school class

GPT Image 2.5 · Text to image · EN

Written like a lesson brief: audience, title, required stages and every molecule to label, plus what to avoid. Use high quality for dense labels. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Dress a person in three reference garments, identity unchanged

Dress a person in three reference garments, identity unchanged

GPT Image 2.5 · Edit your photo · EN

Lists everything about the person that must stay fixed, then allows only the clothing to change and asks the lighting to match. The same pattern keeps a product recognizable. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Farmers market app screen in an iPhone frame

Farmers market app screen in an iPhone frame

GPT Image 2.5 · Text to image · EN

Describes the app as if it already ships: real sections, hierarchy and spacing, no concept-art words. That keeps the result looking like a usable interface. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Festival crowd in Bethel, New York, August 1969

Festival crowd in Bethel, New York, August 1969

GPT Image 2.5 · Text to image · EN

A place and a date are enough for the model to infer clothing, staging and setting. Inspect the result for period accuracy. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Four-panel comic of a pet left home alone

Four-panel comic of a pet left home alone

GPT Image 2.5 · Text to image · EN

One concrete visual beat per panel, written as action and framing, gives a readable comic with a clear setup and punchline. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Holiday collectible toy airplane in retail packaging

Holiday collectible toy airplane in retail packaging

GPT Image 2.5 · Text to image · EN

Uses product photography cues for materials, lighting and label printing, keeps the design original, and fixes the packaging text verbatim. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
How an automatic coffee machine works, as an infographic

How an automatic coffee machine works, as an infographic

GPT Image 2.5 · Text to image · EN

Three plain sentences name the machine, the parts in order and the goal of understanding the flow. The model fills in layout and labels; check the labels for accuracy. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Line drawing of a river valley rendered as a photo

Line drawing of a river valley rendered as a photo

GPT Image 2.5 · Edit your photo · EN

Locks layout, proportions and perspective, lets the model choose materials and light, and bans new elements so it does not reinterpret the sketch. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Market opportunity pitch deck slide with real numbers

Market opportunity pitch deck slide with real numbers

GPT Image 2.5 · Text to image · EN

Written as a slide spec with the actual numbers, chart type, footnotes and logo slot, plus a list of things to avoid. The figures are sample data; replace them before use. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Move a dog from one photo into another

Move a dog from one photo into another

GPT Image 2.5 · Edit your photo · EN

Images are named by number, with what moves, where it goes and what stays unchanged. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Nostalgic Christmas card with an old teddy bear

Nostalgic Christmas card with an old teddy bear

GPT Image 2.5 · Text to image · EN

Scene, mood, style and constraints are labeled blocks, and the card text is given verbatim as the only text allowed. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Pixel art rider in the style of a reference image

Pixel art rider in the style of a reference image

GPT Image 2.5 · Edit your photo · EN

The reference image is used for style only, and the new subject is described separately, so the palette and pixel grid carry over without copying the scene. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Product billboard on a highway at sunset

Product billboard on a highway at sunset

GPT Image 2.5 · Edit your photo · EN

Puts the product from a photo onto a billboard and quotes the copy as exact and verbatim, with typography and a rule that it appears once. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Product cutout on a transparent background

Product cutout on a transparent background

GPT Image 2.5 · Edit your photo · EN

Asks for a crisp silhouette with no halos, exact geometry and legible labels, and bans backdrops and drawn checkerboards. Use PNG or WebP with background=transparent. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Put a person into a dusk campsite chase scene

Put a person into a dusk campsite chase scene

GPT Image 2.5 · Edit your photo · EN

Describes pose, gaze, clothing, place and time of day, and keeps asking for an unstyled real photo instead of a movie poster look. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Remove a flower from a man's hand

Remove a flower from a man's hand

GPT Image 2.5 · Edit your photo · EN

Names the one object to remove and says nothing else may change, which keeps the edit local without a mask. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Same billboard, now on a snowy winter evening

Same billboard, now on a snowy winter evening

GPT Image 2.5 · Edit your photo · EN

The previous output goes back in with one short change. Narrow follow-ups show which change helped. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Same hero rescues a squirrel in the snow

Same hero rescues a squirrel in the snow

GPT Image 2.5 · Edit your photo · EN

Feeds the character image back in and repeats its defining details under Character Consistency, so the outfit and face stay the same in a new scene. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Streetwear campaign shot with one exact tagline

Streetwear campaign shot with one exact tagline

GPT Image 2.5 · Text to image · EN

The tagline is quoted and the prompt asks for it exactly once, then bans extra text, watermarks and unrelated logos. This is the pattern for any ad with fixed copy. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Swap only the chairs in a kitchen photo

Swap only the chairs in a kitchen photo

GPT Image 2.5 · Edit your photo · EN

Writes ONLY in capitals for the one object to replace and lists what to preserve: camera angle, lighting, floor shadows and surrounding objects. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Translate an infographic into Spanish without touching the layout

Translate an infographic into Spanish without touching the layout

GPT Image 2.5 · Edit your photo · EN

A one-line edit on the coffee machine infographic. Check the output for words left in the original language. Preview: the Sunburst variant.

Source: OpenAI image prompting guide
Watercolor hero for a children's book

Watercolor hero for a children's book

GPT Image 2.5 · Text to image · EN

Character, theme, style and constraints are separate labeled blocks, and the plain background shows the character clearly for reuse in later pages. Preview: the Sunburst variant.

Source: OpenAI image prompting guide

Prefer to learn the craft step by step? The 16 chapter course is here.