Examples for using gpt-image-2 through RunAPI from agent tools or API calls. Copy a prompt, then use it in Claude Code, Codex, Cursor, Windsurf, or your backend.
Create a hyper-realistic 3D holographic blueprint projection of a [CAR NAME] emerging as a luminous cyan/blue wireframe model from a dark blue technical blueprint sheet on a wooden workshop table. The car appears as glowing electric blue wireframe lines with grid patterns and transparent surfaces, creating a futuristic holographic effect. The 3D model hovers above the flat blueprint drawings with dramatic lighting and bright cyan glow. Surround with precision drafting tools, compass, rolled blueprints, and technical instruments on dark wooden surface. Dark atmospheric background with studio lighting. Aspect ratio: 1:1 (1080x1080 pixels). Do not include any vehicle logos or brand emblems
curl -X POST https://runapi.ai/api/v1/gpt_image_2/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "gpt-image-2",
"prompt": "Create a hyper-realistic 3D holographic blueprint projection of a [CAR NAME] emerging as a luminous cyan/blue wireframe model from a dark blue technical blueprint sheet on a wooden workshop table. The car appears as glowing electric blue wireframe lines with grid patterns and transparent surfaces, creating a futuristic holographic effect. The 3D model hovers above the flat blueprint drawings with dramatic lighting and bright cyan glow. Surround with precision drafting tools, compass, rolled blueprints, and technical instruments on dark wooden surface. Dark atmospheric background with studio lighting. Aspect ratio: 1:1 (1080x1080 pixels). Do not include any vehicle logos or brand emblems"
}
JSON
A charcuterie-style board presents JAPAN as an edible sculpture: • Letters carved from compact sushi-rice blocks wrapped with nori trim • Rivers and lakes painted in glossy soy-mirin reduction • Mountain ranges piped in smooth wasabi-avocado mousse (Mount Fuji peak highlighted) • Coastlines dusted with shichimi togarashi spice mix Add one iconic red Torii-gate bamboo skewer garnish anchoring the composition. Polarised studio softbox lighting, 50 mm prime, f/4, shallow depth; crumbs, knife marks, and wood-grain preserved. Rendered in Cinema 4D + Octane, 16 : 9 warm tabletop scene. (NO text, flags, or extra objects outside the board.)
curl -X POST https://runapi.ai/api/v1/gpt_image_2/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "gpt-image-2",
"prompt": "A charcuterie-style board presents JAPAN as an edible sculpture: • Letters carved from compact sushi-rice blocks wrapped with nori trim • Rivers and lakes painted in glossy soy-mirin reduction • Mountain ranges piped in smooth wasabi-avocado mousse (Mount Fuji peak highlighted) • Coastlines dusted with shichimi togarashi spice mix Add one iconic red Torii-gate bamboo skewer garnish anchoring the composition. Polarised studio softbox lighting, 50 mm prime, f/4, shallow depth; crumbs, knife marks, and wood-grain preserved. Rendered in Cinema 4D + Octane, 16 : 9 warm tabletop scene. (NO text, flags, or extra objects outside the board.)"
}
JSON
A photorealistic image, split down the middle. On the left side, a small, detailed bone that touches the center divider, resembling a dog bone set against a clear blue sky with clouds in background. On the right side, the bone seamlessly transitions into a detailed, futuristic stunning spacecraft, maintaining the line and shape of the bone against a deep space background scattered with stars. The spacecraft extends to the right edge, designed with elements that suggest a continuation of the bone's structure. The transition should be smooth, illustrating a creative blend where the organic shape of the bone becomes one with the mechanical structure of the spacecraft, set against a neutral background to emphasize the two subjects.
curl -X POST https://runapi.ai/api/v1/gpt_image_2/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "gpt-image-2",
"prompt": "A photorealistic image, split down the middle. On the left side, a small, detailed bone that touches the center divider, resembling a dog bone set against a clear blue sky with clouds in background. On the right side, the bone seamlessly transitions into a detailed, futuristic stunning spacecraft, maintaining the line and shape of the bone against a deep space background scattered with stars. The spacecraft extends to the right edge, designed with elements that suggest a continuation of the bone's structure. The transition should be smooth, illustrating a creative blend where the organic shape of the bone becomes one with the mechanical structure of the spacecraft, set against a neutral background to emphasize the two subjects."
}
JSON
A full-body 3D stylized cartoon boy, Pixar/Disney-inspired character design, cute and modern look, generated from the provided reference photo. Preserve the subject’s facial structure, proportions, and identity while translating them into a stylized animated form. He has well-groomed short-to-medium hair styled to match the reference, expressive eyes, soft facial features, and a friendly, confident smile. He is wearing a clean white oversized t-shirt, a high-waisted casual jacket or hoodie (optional), slim-fit pants, and modern sneakers that match the color palette of the outfit. A small crossbody sling bag or backpack adds a trendy touch. Slim proportions, smooth skin, polished 3D render, standing in a relaxed walking or casual pose. Clean white background, soft studio lighting, high detail, ultra-smooth textures, vibrant yet natural colors, modern fashion doll aesthetic, professional character render, cinematic quality, 4K resolution.
curl -X POST https://runapi.ai/api/v1/gpt_image_2/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "gpt-image-2",
"prompt": "A full-body 3D stylized cartoon boy, Pixar/Disney-inspired character design, cute and modern look, generated from the provided reference photo. Preserve the subject’s facial structure, proportions, and identity while translating them into a stylized animated form. He has well-groomed short-to-medium hair styled to match the reference, expressive eyes, soft facial features, and a friendly, confident smile. He is wearing a clean white oversized t-shirt, a high-waisted casual jacket or hoodie (optional), slim-fit pants, and modern sneakers that match the color palette of the outfit. A small crossbody sling bag or backpack adds a trendy touch. Slim proportions, smooth skin, polished 3D render, standing in a relaxed walking or casual pose. Clean white background, soft studio lighting, high detail, ultra-smooth textures, vibrant yet natural colors, modern fashion doll aesthetic, professional character render, cinematic quality, 4K resolution."
}
JSON
Create a highly detailed, hyper-realistic image of a miniature diorama inside a glass dome, placed on a wooden table near a softly lit window. Inside the dome, show a [COUNTRY] cultural figure with iconic attire, prominently positioned in the center foreground taking up significant space in the composition to really “pop” from the scene, smiling and waving, in a stylized 3D miniature version of [COUNTRY]’s iconic setting. Include the most iconic background features of that location, all scaled in a whimsical, miniature style. The time of day, season, and mood will vary by [COUNTRY]’s cultural theme. The scene is captured from a slightly elevated perspective. The base of the dome has a wooden plaque with bold, gold-colored text labeled: ‘[COUNTRY]’ in a strong, readable font that stands out prominently. The scene should be bright, magical, joyful, and feel like a collectible souvenir moment from a dream trip. Aspect ratio: 1080x1080
curl -X POST https://runapi.ai/api/v1/gpt_image_2/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "gpt-image-2",
"prompt": "Create a highly detailed, hyper-realistic image of a miniature diorama inside a glass dome, placed on a wooden table near a softly lit window. Inside the dome, show a [COUNTRY] cultural figure with iconic attire, prominently positioned in the center foreground taking up significant space in the composition to really “pop” from the scene, smiling and waving, in a stylized 3D miniature version of [COUNTRY]’s iconic setting. Include the most iconic background features of that location, all scaled in a whimsical, miniature style. The time of day, season, and mood will vary by [COUNTRY]’s cultural theme. The scene is captured from a slightly elevated perspective. The base of the dome has a wooden plaque with bold, gold-colored text labeled: ‘[COUNTRY]’ in a strong, readable font that stands out prominently. The scene should be bright, magical, joyful, and feel like a collectible souvenir moment from a dream trip. Aspect ratio: 1080x1080"
}
JSON
Eye-level straight-on view, looking directly into a hyper-vibrant 3D layered paper cut-out diorama. Pokéball red and white, Pikachu yellow, sky blue and grass green accents, extremely high saturation, PHYSICAL PAPER LAYERS clearly visible with hard cut edges and thick paper cross-sections at every depth plane. Maximalist composition, crowded Pokémon world scene — layered Pokémon Center and gym architecture receding through all depth planes, Pikachu as central figure mid-thunderbolt on foreground layer, dozens of iconic Pokémon — Charizard, Gengar, Eevee, Snorlax, Mewtwo — each inhabiting separate depth planes, trainer Ash and rivals threading between battle layers, Pokéball burst effects frozen mid-release across all planes, tall grass and route pathways between habitat layers, legendary Pokémon silhouettes in remote background layers, scattered Pokéballs, badges and berry items crowding the vibrant foreground. Soft single-direction studio lighting creating deep drop shadows between every physical paper layer, matte paper texture throughout, NO glow, NO soft edges, NO volumetric lighting, octane render, 8k, clean details --ar 3:4
curl -X POST https://runapi.ai/api/v1/gpt_image_2/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "gpt-image-2",
"prompt": "Eye-level straight-on view, looking directly into a hyper-vibrant 3D layered paper cut-out diorama. Pokéball red and white, Pikachu yellow, sky blue and grass green accents, extremely high saturation, PHYSICAL PAPER LAYERS clearly visible with hard cut edges and thick paper cross-sections at every depth plane. Maximalist composition, crowded Pokémon world scene — layered Pokémon Center and gym architecture receding through all depth planes, Pikachu as central figure mid-thunderbolt on foreground layer, dozens of iconic Pokémon — Charizard, Gengar, Eevee, Snorlax, Mewtwo — each inhabiting separate depth planes, trainer Ash and rivals threading between battle layers, Pokéball burst effects frozen mid-release across all planes, tall grass and route pathways between habitat layers, legendary Pokémon silhouettes in remote background layers, scattered Pokéballs, badges and berry items crowding the vibrant foreground. Soft single-direction studio lighting creating deep drop shadows between every physical paper layer, matte paper texture throughout, NO glow, NO soft edges, NO volumetric lighting, octane render, 8k, clean details --ar 3:4"
}
JSON
A square, 1:1 photorealistic close-up photograph of an incredibly detailed physical miniature geography and city model of [CITY], focused only on the city’s most recognizable landmark area (not the full city map). The scene shows a compact 3D topographic environment built around the iconic landmark zone — realistic terrain, elevation changes, nearby rivers or coastline if relevant, and surrounding urban density scaled naturally to highlight the landmark as the clear focal point. The model feels like a premium museum diorama rather than a map. Thousands of tiny, highly detailed buildings, roads, bridges, and infrastructure cluster organically around the landmark. The landmark itself rises prominently above everything else, instantly recognizable even at a glance. Strong natural side sunlight casts long, realistic shadows across the miniature terrain, enhancing depth and realism. Shallow depth of field (tilt-shift style) keeps the landmark zone sharply in focus while softly blurring the outer edges of the square frame. The entire scene appears as a tangible, physical model illuminated by real-world light. Above the city, the sky is perfectly clean and empty — no clouds at all — except for the city name. The name [CITY] appears integrated into the sky as thick, three-dimensional cloud typography, sculpted from dense white vapor. The letters are bold, volumetric, and clearly readable, with soft shadowing and subtle depth, floating naturally in the open sky. Opacity ~70%, but with enough thickness to remain instantly legible. A small, minimal national flag is placed subtly near the sky text (corner or aligned to the typography), balanced and unobtrusive. Format: Square 1080×1080 Style: Ultra-photorealistic, cinematic, premium miniature realism Mood: Awe, clarity, instantly recognizable, highly shareable
curl -X POST https://runapi.ai/api/v1/gpt_image_2/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "gpt-image-2",
"prompt": "A square, 1:1 photorealistic close-up photograph of an incredibly detailed physical miniature geography and city model of [CITY], focused only on the city’s most recognizable landmark area (not the full city map). The scene shows a compact 3D topographic environment built around the iconic landmark zone — realistic terrain, elevation changes, nearby rivers or coastline if relevant, and surrounding urban density scaled naturally to highlight the landmark as the clear focal point. The model feels like a premium museum diorama rather than a map. Thousands of tiny, highly detailed buildings, roads, bridges, and infrastructure cluster organically around the landmark. The landmark itself rises prominently above everything else, instantly recognizable even at a glance. Strong natural side sunlight casts long, realistic shadows across the miniature terrain, enhancing depth and realism. Shallow depth of field (tilt-shift style) keeps the landmark zone sharply in focus while softly blurring the outer edges of the square frame. The entire scene appears as a tangible, physical model illuminated by real-world light. Above the city, the sky is perfectly clean and empty — no clouds at all — except for the city name. The name [CITY] appears integrated into the sky as thick, three-dimensional cloud typography, sculpted from dense white vapor. The letters are bold, volumetric, and clearly readable, with soft shadowing and subtle depth, floating naturally in the open sky. Opacity ~70%, but with enough thickness to remain instantly legible. A small, minimal national flag is placed subtly near the sky text (corner or aligned to the typography), balanced and unobtrusive. Format: Square 1080×1080 Style: Ultra-photorealistic, cinematic, premium miniature realism Mood: Awe, clarity, instantly recognizable, highly shareable"
}
JSON
Do this for Pineapple Pizza: <instruction> Input A: user uploads an image or shares name of dish Logic Identify the historical inventor (e.g., Raffaele Esposito or Henri Charpentier) and the exact year of origin. Task: A hyper-realistic 4:5 macro photograph of an oversized, open antique culinary codex resting on a dark velvet museum plinth. Left Page (The Living Diorama): The left side of the book is hollowed out like a secret compartment. Inside is a breathtaking 3D miniature scene. A highly detailed figurine of the dish’s inventor is captured mid-motion in a period-accurate kitchen. Around them are microscopic versions of the 10-15 key ingredients, each in its own tiny hand-blown glass vial or micro-wooden crate. Include miniature brass cooking tools specific to the era. The scene is lit from within the "pages" by a warm, magical amber glow. Right Page (The Technical Recipe): The right page is flat, aged parchment featuring elegant, faded Spencerian calligraphy and hand-painted watercolor illustrations. 1. Top: The dish name in both English and its native language, with the bold "Origin Date." 2. Middle: A vertical "Ingredient Blueprint" with hyper-detailed sketches of each raw component. 3. Bottom: A small, detailed "Origin Map" showing the specific city of birth, styled like a 19th-century cartographic inset. 4. Text: Visible, legible recipe steps written in ink that looks slightly raised on the paper. Style: Museum specimen photography. 85mm macro lens. The lighting should be a mix of cool gallery spotlights and the warm "internal" glow of the book's diorama. Extreme texture on the weathered leather binding and the tooth of the paper. Output: ONE image, 4:5 aspect ratio. </instruction>
curl -X POST https://runapi.ai/api/v1/gpt_image_2/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "gpt-image-2",
"prompt": "Do this for Pineapple Pizza: <instruction> Input A: user uploads an image or shares name of dish Logic Identify the historical inventor (e.g., Raffaele Esposito or Henri Charpentier) and the exact year of origin. Task: A hyper-realistic 4:5 macro photograph of an oversized, open antique culinary codex resting on a dark velvet museum plinth. Left Page (The Living Diorama): The left side of the book is hollowed out like a secret compartment. Inside is a breathtaking 3D miniature scene. A highly detailed figurine of the dish’s inventor is captured mid-motion in a period-accurate kitchen. Around them are microscopic versions of the 10-15 key ingredients, each in its own tiny hand-blown glass vial or micro-wooden crate. Include miniature brass cooking tools specific to the era. The scene is lit from within the \"pages\" by a warm, magical amber glow. Right Page (The Technical Recipe): The right page is flat, aged parchment featuring elegant, faded Spencerian calligraphy and hand-painted watercolor illustrations. 1. Top: The dish name in both English and its native language, with the bold \"Origin Date.\" 2. Middle: A vertical \"Ingredient Blueprint\" with hyper-detailed sketches of each raw component. 3. Bottom: A small, detailed \"Origin Map\" showing the specific city of birth, styled like a 19th-century cartographic inset. 4. Text: Visible, legible recipe steps written in ink that looks slightly raised on the paper. Style: Museum specimen photography. 85mm macro lens. The lighting should be a mix of cool gallery spotlights and the warm \"internal\" glow of the book's diorama. Extreme texture on the weathered leather binding and the tooth of the paper. Output: ONE image, 4:5 aspect ratio. </instruction>"
}
JSON
Surreal diorama emerging from an open vintage book of "Brave New World" by Aldous Huxley: the pages form a futuristic hatching laboratory with rows of glowing test tubes containing embryonic figures on conveyor belts, lab-coated workers in white uniforms operating machines, a large all-seeing eye symbol on the wall, conveyor belts moving via hidden mechanisms, cool blue neon lighting, hyper-detailed miniature sculpture, clinical and eerie atmosphere, photorealistic.
curl -X POST https://runapi.ai/api/v1/gpt_image_2/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "gpt-image-2",
"prompt": "Surreal diorama emerging from an open vintage book of \"Brave New World\" by Aldous Huxley: the pages form a futuristic hatching laboratory with rows of glowing test tubes containing embryonic figures on conveyor belts, lab-coated workers in white uniforms operating machines, a large all-seeing eye symbol on the wall, conveyor belts moving via hidden mechanisms, cool blue neon lighting, hyper-detailed miniature sculpture, clinical and eerie atmosphere, photorealistic."
}
JSON
2x2 grid, do this for 4 hidden gem international foods select * from popup_recipe_visualizer where recipe_target = '[dish_name]' and structural_base = 'l-shaped_open_box_white_interior' and wall_layer_2d = ( select line_art, typography, botanical_drawings from ingredient_inference_db where dish = '[dish_name]' ) and foreground_layer_3d = array[ 'bowl_of_completed_[dish_name]', 'kawaii_chibi_mascot_interacting_with_food', 'scattered_3d_garnishes_on_floor' ] and dynamic_action = 'thick_glossy_sauce_pouring_from_2d_wall_hole_into_3d_bowl' and render_engine = 'octane_render_soft_studio_lighting';
%{model} is available through RunAPI as part of the unified model catalog. These prompts show practical input patterns that agents and backend services can reuse.
How do I use these prompts?
Copy any prompt and paste it into Claude Code, Codex, Cursor, or Windsurf after installing the RunAPI MCP Server. Developers can also copy the API example and send the prompt directly.
Do these prompts cost money to browse?
Browsing and copying prompt examples is free. Generation requests only cost money when you call a RunAPI model with your API key.
Can I adapt the prompts for production?
Yes. Treat each prompt as a starting point, then add your brand rules, output dimensions, safety constraints, and application-specific context before using it in production.