3:2
#1
PhotorealismBest on FLUX.1 Pro
A weathered fisherman mending nets at golden hour on a Portuguese dock, shot on Hasselblad X2D, 80mm f/2.5, soft rim light, fine grain, Kodak Portra 400 color science.
3:2
TL;DR. This page is the canonical ZeroTwo library for both ai image prompt (singular) and ai image prompts (plural) queries — 50 ready-to-copy AI image prompts split across six categories, each tagged with the model it renders best on. Copy any AI image prompt, paste it into the ZeroTwo image studio, and run it across DALL-E 3, FLUX.1, SDXL, Recraft, and Ideogram in one tab. The remix framework and cross-model matrix below turn one prompt into a dozen variations in minutes.
The 50-prompt gallery
Filter by category. Every card includes the full AI image prompt, the recommended model, an aspect-ratio tag, and a one-tap copy button.
- 16:9#2PhotorealismBest on Imagen 3
Tokyo street market at dusk, neon kanji reflected in rain puddles, candid pedestrians, Sony A7R V, 35mm f/1.4, ISO 800, cinematic teal-and-amber grade.
16:9 - 3:2#3PhotorealismBest on FLUX.1 Pro
Snow leopard mid-leap across a Himalayan ridge at sunrise, 600mm f/4 telephoto, frozen motion, dust crystals catching light, National Geographic style.
3:2 - 1:1#4PhotorealismBest on DALL-E 3
Elderly woman laughing in a sunlit Marrakech kitchen, flour on her hands, shot on Leica Q3, 28mm f/1.7, warm tungsten fill, natural skin texture.
1:1 - 16:9#5PhotorealismBest on FLUX.1 Pro
Abandoned brutalist library reclaimed by ferns, shafts of dust-lit sun through broken skylights, full-frame medium-format, deep focus, muted cyan grade.
16:9 - 1:1#6PhotorealismBest on Imagen 3
Chef plating a single scallop with foam, overhead shot, marble surface, soft north-window light, food editorial style for Bon Appetit.
1:1 - 16:9#7PhotorealismBest on FLUX.1 Pro
Lone surfer paddling out into a glassy dawn swell at Trestles, water droplets frozen by 1/2000s shutter, Canon R5, 100-500mm at 400mm, salt haze.
16:9 - 3:2#8PhotorealismBest on Imagen 3
Architectural portrait of a mid-century Palm Springs house at twilight, blue-hour sky, warm interior lighting, Tilt-shift lens, Julius Shulman tribute.
3:2 - 1:1#9PhotorealismBest on DALL-E 3
Street portrait of a Berlin musician with violet hair, harsh midday sun, square crop, Fujifilm GFX, environmental detail, no retouching.
1:1 - 3:4#10IllustrationBest on Recraft V3
Editorial illustration of a person climbing a staircase made of open books, flat shapes, limited palette of cobalt and ochre, halftone shadow texture, New Yorker style.
3:4 - 4:5#11IllustrationBest on Midjourney v7
Children's book illustration of a sleepy fox curled in a teacup garden, watercolor wash, soft graphite outlines, pastel meadow palette.
4:5 - 2:3#12IllustrationBest on Recraft V3
Bold vector poster of a cyclist racing through a Tour de France mountain stage, art-deco geometric forms, four-color screenprint feel, AIGA award style.
2:3 - 16:9#13IllustrationBest on Recraft V3
Risograph-style spread of a busy farmers market, pink and teal duotone, intentional registration offset, flat people with simple linework.
16:9 - 3:4#14IllustrationBest on Midjourney v7
Botanical scientific illustration of a Venus flytrap with cross-section, ink and gouache on warm paper, neat handwritten Latin labels, 1850s naturalist style.
3:4 - 2:3#15IllustrationBest on Midjourney v7
Comic-book splash panel of a hero leaping between rooftops at night, dynamic ink lines, Ben-Day dots, two-tone red and cyan palette.
2:3 - 16:10#16IllustrationBest on Recraft V3
Minimal flat-design illustration of a remote-work setup at sunrise, single-line desk plant, three-color palette, generous negative space, dribbble-ready.
16:10 - 2:3#17IllustrationBest on Midjourney v7
Tarot-card style illustration of 'The Coder', glowing keyboard halo, cosmic sigils, gold-leaf accents on midnight blue, art-nouveau border.
2:3 - 1:1#183D RenderBest on FLUX.1 Pro
Octane render of a glossy ceramic bunny on a pastel pedestal, soft three-point studio lighting, subsurface scattering, 8k, isometric framing.
1:1 - 16:9#193D RenderBest on FLUX.1 Pro
Cyberpunk hover-bike concept, Blender Cycles render, brushed-aluminum panels, neon underglow, volumetric fog, studio HDRI, sci-fi product shot.
16:9 - 1:1#203D RenderBest on Recraft V3
Low-poly diorama of a cozy mountain cabin in winter, pastel snow, oversized pine trees, soft ambient occlusion, mobile-game art-direction.
1:1 - 1:1#213D RenderBest on FLUX.1 Pro
Photoreal 3D of melting chocolate sphere on a marble plinth, caustics through the chocolate puddle, macro lens, dramatic side key light.
1:1 - 1:1#223D RenderBest on Recraft V3
Stylized voxel city floating on a sky island, waterfalls cascading off the edge, soft cel shading, isometric camera, Monument-Valley palette.
1:1 - 16:9#233D RenderBest on Imagen 3
Architectural visualization of a cantilevered glass house over a forest stream, dusk lighting, accurate IES lights, V-Ray realism, drone perspective.
16:9 - 16:9#243D RenderBest on Recraft V3
Inflatable typography spelling 'PLAY' in glossy hot-pink latex, soft shadows on a cream cyclorama, Cinema 4D Redshift, fashion-editorial palette.
16:9 - 3:4#253D RenderBest on FLUX.1 Pro
Hyperreal 3D character bust of an elderly samurai, pore-level skin detail, weathered armor metalwork, Substance Painter textures, ZBrush sculpt.
3:4 - 16:9#26AnimeBest on SDXL (Animagine)
Anime girl in a yukata watching summer fireworks from a wooden footbridge, Makoto Shinkai style, hyper-saturated sky, painterly cloud detail, lens flare.
16:9 - 16:9#27AnimeBest on Midjourney v7 (niji)
Studio Ghibli inspired forest spirit walking through a moss kingdom, soft watercolor backgrounds, Joe Hisaishi mood, warm filtered sunlight.
16:9 - 4:3#28AnimeBest on SDXL (Animagine)
90s mecha pilot in cockpit, glowing CRT readouts, halftone shadows, cel-shaded ink lines, Patlabor color palette, side-profile composition.
4:3 - 1.85:1#29AnimeBest on Midjourney v7 (niji)
Slice-of-life cafe scene at twilight, warm window light on a barista, soft shoujo art style, pastel rim light, cinematic 1.85:1 framing.
1.85:1 - 2:3#30AnimeBest on SDXL (Animagine)
Dark-fantasy anime swordsman silhouetted against a blood-red moon, ink-spatter background, manga screentone shading, chiaroscuro lighting.
2:3 - 2:3#31AnimeBest on Midjourney v7 (niji)
Magical-girl transformation pose, sparkles, ribbon flourishes, vivid cyan-magenta palette, classic Sailor Moon era cel art, full-body shot.
2:3 - 21:9#32AnimeBest on SDXL (Animagine)
Cyberpunk anime cityscape at 3am, holographic billboards in Japanese, lone figure on a rooftop, Akira-era animation cels, deep magenta neon.
21:9 - 1:1#33AnimeBest on Midjourney v7 (niji)
Chibi-style tiny dragon napping on a stack of pancakes, oversized eyes, soft pastel palette, sticker-ready white outline.
1:1 - 1:1#34AbstractBest on FLUX.1 Pro
Liquid chrome ribbons curling through a void, iridescent oil-slick highlights, ultra-glossy material, studio HDRI, generative-art aesthetic.
1:1 - 2:3#35AbstractBest on Recraft V3
Geometric Bauhaus composition of overlapping circles and triangles, primary palette of red, yellow, blue, riso-print texture, museum-poster scale.
2:3 - 1:1#36AbstractBest on FLUX.1 Pro
Macro photo of swirling alcohol-ink on yupo paper, bloom of magenta into amber, organic edge detail, abstract expressionist energy.
1:1 - 16:9#37AbstractBest on SDXL
Generative wave-function art, thousands of thin lines morphing across a black canvas, neon mint accents, Processing-sketch aesthetic.
16:9 - 3:4#38AbstractBest on Ideogram 2.0
Folded paper sculpture in pastel gradient, soft top-down studio light, minimal shadow, zen-like composition, suitable for editorial cover.
3:4 - 2:3#39AbstractBest on Midjourney v7
Fractured stained-glass mosaic of a phoenix in flight, warm reds and gold leaf, hand-leaded outlines, light streaming through composition.
2:3 - 1:1#40AbstractBest on Recraft V3
Topographic-line abstract of a fictional island, contour lines in deep navy on cream, subtle altitude shading, cartographic art print.
1:1 - 9:16#41AbstractBest on FLUX.1 Pro
Smoke rising in slow motion through a single colored beam of light, photographed at 1/4000s, deep black background, fine particulate detail.
9:16 - 4:5#42Product PhotographyBest on FLUX.1 Pro
Premium amber glass perfume bottle on rippled black sand, soft top key + warm side rim, condensation droplets, hero e-commerce angle.
4:5 - 1:1#43Product PhotographyBest on Imagen 3
Minimalist running shoe floating against a sage backdrop, soft cyclorama, single soft light, subtle ground shadow, Nike-style hero crop.
1:1 - 4:5#44Product PhotographyBest on FLUX.1 Pro
Skincare jar surrounded by frosted ice cubes and eucalyptus, beaded condensation, crisp top-down macro, magazine-spread composition.
4:5 - 4:5#45Product PhotographyBest on DALL-E 3
Matte ceramic coffee mug on warm oak, single shaft of morning window light, wisp of steam, lifestyle product photography for a roaster brand.
4:5 - 1:1#46Product PhotographyBest on FLUX.1 Pro
Wireless earbuds case shot in mid-splash of clear water, frozen droplets, dark gradient backdrop, hero tech product image.
1:1 - 4:5#47Product PhotographyBest on Imagen 3
Hand-poured candle in a glass vessel, wick lit, soft golden halo, bookshelf bokeh background, hygge editorial mood, 50mm f/1.8.
4:5 - 2:3#48Product PhotographyBest on FLUX.1 Pro
Athletic protein-shake bottle suspended in mid-air with powder cloud burst, dramatic side-strobe, deep navy seamless, motion-frozen detail.
2:3 - 4:5#49Product PhotographyBest on DALL-E 3
Boxed gourmet chocolates with one piece bitten, soft window light, linen surface, artisanal brand storytelling shot, shallow depth of field.
4:5 - 1:1#50Product PhotographyBest on Ideogram 2.0
Stack of three cosmetic compacts on a blush pink riser, geometric backdrop with playful drop shadows, color-blocked beauty editorial.
1:1
The remix framework: 4 swap axes
One AI image prompt becomes a dozen variations by swapping just four things. This is the same axis system Midjourney educator Nick St. Pierre teaches in his prompt workshops: subject, style, lighting, aspect ratio.
Subject
fisherman → blacksmith → falconer → cellist
Style
Hasselblad photo → ink illustration → 3D render → anime cel
Lighting
golden hour → blue hour → harsh noon → studio strobe
Aspect ratio
1:1 (social) → 16:9 (cinema) → 9:16 (story) → 2:3 (print)
Worked example. Start from prompt #1 (fisherman, Hasselblad photo, golden hour, 3:2). Swap to cellist, ink illustration, blue hour, 9:16 — same compositional bones, entirely different image. Repeat the swap on a second model and you now have eight reference frames from one starting prompt.
DALL-E 3
Substantially better at prompt following than DALL-E 2 per OpenAI's system card.
Source: OpenAI system cardRun any prompt across 8+ image generators
One tab, one input, eight outputs. $19.99/mo.
Cross-model compatibility matrix
Every prompt runs everywhere — but the result varies. This matrix shows which AI image prompt category lands best on which generator.
| Style | Best on | Also strong | Notes |
|---|---|---|---|
| Photorealism | FLUX.1 Pro | Imagen 3 | FLUX leads on prompt fidelity; Imagen 3 wins on text-in-image. |
| Illustration | Recraft V3 | Midjourney v7 | Recraft is vector-aware; Midjourney has the painterly polish. |
| 3D Render | FLUX.1 Pro | Recraft V3 | FLUX nails materials; Recraft handles low-poly stylization well. |
| Anime | Midjourney v7 (niji) | SDXL (Animagine) | Niji owns shoujo/Ghibli vibes; SDXL fine-tunes excel at manga lines. |
| Abstract | FLUX.1 Pro | SDXL | FLUX for liquid/chrome; SDXL for generative line art. |
| Product | FLUX.1 Pro | Imagen 3 | FLUX for hero shots; Imagen 3 for accurate brand text on packaging. |
"Subject, medium, style, lighting, composition. Five slots. If your prompt has all five, you're already ahead of 95% of the people generating images today."
Frequently asked questions
What is an AI image prompt?
An AI image prompt is the text instruction you give a generative image model to produce an image. The strongest prompts name a subject, a medium, a style reference, lighting, and composition — five slots that account for most of the visible difference between a mediocre output and a great one.
Are these AI image prompts free to copy?
Yes. Every prompt on this page is free to copy, modify, and use commercially. The text is the input — what the image generator returns is governed by the model's terms of service, which vary by provider.
Which AI gives the best image from this prompt?
It depends on the style. FLUX.1 Pro currently leads the Artificial Analysis text-to-image arena for photoreal and 3D, Midjourney v7 wins on painterly illustration and anime, Imagen 3 dominates text-in-image, Recraft V3 is the only vector-native generator. Run the same prompt across all of them in the ZeroTwo image studio to compare outputs side by side.
How do I write my own AI image prompts?
Use a five-slot formula: subject, medium, style reference, lighting, composition. Add aspect ratio and a model-specific quality cue last. Our companion guide on how to write image prompts walks through the seven-slot expansion with worked examples.
What is the difference between this page and how-to-write-image-prompts?
This page is a curated library — 50 ready-to-paste prompts you can use right now. The how-to page is a tutorial that teaches you to write your own from scratch. Most users start here, then graduate to the tutorial when they want custom outputs.
Can I use these AI image prompts on DALL-E, Midjourney, and Flux?
Yes. The prompts are written in plain natural language and work on every major generator. We tag each prompt with a recommended model based on which one produces the strongest result, but every prompt will run anywhere.
How long should an AI image prompt be?
Typically 25 to 60 words. Shorter prompts give the model freedom and produce surprising results; longer prompts give you control. Most prompts on this page sit in the 30 to 50 word range — long enough to be specific, short enough to stay coherent.
Why does the same prompt look different on different models?
Each model has its own training data, default style, and prompt parser. FLUX is more literal, Midjourney aestheticizes aggressively, SDXL fine-tunes specialize per genre. Running the same prompt on all of them reveals the difference instantly.
Key takeaways
- 50 curated AI image prompts across 6 categories — copy-ready, no signup needed.
- Every prompt is tagged with the model that renders it best.
- The remix framework turns one prompt into a dozen variations in minutes.
- The same prompt looks meaningfully different across FLUX, Midjourney, DALL-E, SDXL, Imagen, Recraft, and Ideogram — run them side by side.
- ZeroTwo gives you 8+ image generators in one tab for $19.99/mo.
Keep going
- How to Write Image PromptsThe companion tutorial — learn the 7-slot prompt formula.
- Best AI Image GeneratorTop 8 image generators ranked by arena Elo and license freedom.
- AI Image Generator from TextRun text prompts through every major model in one place.
- Anime AI Art PromptsA focused library for anime, manga, and Ghibli-style outputs.
- Prompt EnhancerAuto-rewrite your draft prompt into a model-tuned final version.
ZeroTwo Editorial — Image Models Team
We test every frontier image generator weekly across the same 50-prompt test set you see on this page. Recommendations reflect Apr–May 2026 outputs on FLUX.1 Pro, Midjourney v7, DALL-E 3, Imagen 3, SDXL, Recraft V3, and Ideogram 2.0.
Published · Last updated
Stop bouncing between tabs.
Run any AI image prompt across DALL-E 3, FLUX.1, SDXL, Recraft, Ideogram, and more — in one tab, on one bill.
Start with ZeroTwo — $19.99/mo