Examples for using nano-banana-2 through RunAPI from agent tools or API calls. Copy a prompt, then use it in Claude Code, Codex, Cursor, Windsurf, or your backend.
1. claude mcp add runapi -s user -- npx -y @runapi.ai/mcp
2. Restart Claude Code
3. Paste this prompt: Generate an image: "Ultra-realistic 8K full body portrait of [PERSON’S FULL NAME], wearing a clean and pressed white social shirt with folded collar and a small lapel microphone, dark navy-blue dress pants and polished brown social shoes. Casually and unpretentiously leaning against a smooth light gray studio wall; hands are in pockets and one leg is crossed over the other, with relaxed and confident body language. Add to the wall next to them a prominent vector portrait in black and white of their face and bust - with sharp lines and angles, overlapping polygonal shapes and a minimalist modern graphic style, right below the information: “[PERSON’S FULL NAME]”, and below the name: “[PROFESSION]”"
1. codex plugin install runapi-mcp@agents
2. Restart Codex
3. Paste this prompt: Generate an image: "Ultra-realistic 8K full body portrait of [PERSON’S FULL NAME], wearing a clean and pressed white social shirt with folded collar and a small lapel microphone, dark navy-blue dress pants and polished brown social shoes. Casually and unpretentiously leaning against a smooth light gray studio wall; hands are in pockets and one leg is crossed over the other, with relaxed and confident body language. Add to the wall next to them a prominent vector portrait in black and white of their face and bust - with sharp lines and angles, overlapping polygonal shapes and a minimalist modern graphic style, right below the information: “[PERSON’S FULL NAME]”, and below the name: “[PROFESSION]”"
1. npx @runapi.ai/mcp init cursor
2. Restart Cursor
3. Paste this prompt: Generate an image: "Ultra-realistic 8K full body portrait of [PERSON’S FULL NAME], wearing a clean and pressed white social shirt with folded collar and a small lapel microphone, dark navy-blue dress pants and polished brown social shoes. Casually and unpretentiously leaning against a smooth light gray studio wall; hands are in pockets and one leg is crossed over the other, with relaxed and confident body language. Add to the wall next to them a prominent vector portrait in black and white of their face and bust - with sharp lines and angles, overlapping polygonal shapes and a minimalist modern graphic style, right below the information: “[PERSON’S FULL NAME]”, and below the name: “[PROFESSION]”"
1. npx @runapi.ai/mcp init windsurf
2. Restart Windsurf
3. Paste this prompt: Generate an image: "Ultra-realistic 8K full body portrait of [PERSON’S FULL NAME], wearing a clean and pressed white social shirt with folded collar and a small lapel microphone, dark navy-blue dress pants and polished brown social shoes. Casually and unpretentiously leaning against a smooth light gray studio wall; hands are in pockets and one leg is crossed over the other, with relaxed and confident body language. Add to the wall next to them a prominent vector portrait in black and white of their face and bust - with sharp lines and angles, overlapping polygonal shapes and a minimalist modern graphic style, right below the information: “[PERSON’S FULL NAME]”, and below the name: “[PROFESSION]”"
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "Ultra-realistic 8K full body portrait of [PERSON’S FULL NAME], wearing a clean and pressed white social shirt with folded collar and a small lapel microphone, dark navy-blue dress pants and polished brown social shoes. Casually and unpretentiously leaning against a smooth light gray studio wall; hands are in pockets and one leg is crossed over the other, with relaxed and confident body language. Add to the wall next to them a prominent vector portrait in black and white of their face and bust - with sharp lines and angles, overlapping polygonal shapes and a minimalist modern graphic style, right below the information: “[PERSON’S FULL NAME]”, and below the name: “[PROFESSION]”"
}
JSON
import { NanoBananaClient } from "@runapi.ai/nano-banana";
const client = new NanoBananaClient({
apiKey: process.env.RUNAPI_API_KEY,
});
const result = await client.textToImage.run({
"model": "nano-banana-2",
"prompt": "Ultra-realistic 8K full body portrait of [PERSON’S FULL NAME], wearing a clean and pressed white social shirt with folded collar and a small lapel microphone, dark navy-blue dress pants and polished brown social shoes. Casually and unpretentiously leaning against a smooth light gray studio wall; hands are in pockets and one leg is crossed over the other, with relaxed and confident body language. Add to the wall next to them a prominent vector portrait in black and white of their face and bust - with sharp lines and angles, overlapping polygonal shapes and a minimalist modern graphic style, right below the information: “[PERSON’S FULL NAME]”, and below the name: “[PROFESSION]”"
});
console.log(result.id);
require "runapi/nano_banana"
client = RunApi::NanoBanana::Client.new
result = client.text_to_image.run(
model: "nano-banana-2",
prompt: "Ultra-realistic 8K full body portrait of [PERSON’S FULL NAME], wearing a clean and pressed white social shirt with folded collar and a small lapel microphone, dark navy-blue dress pants and polished brown social shoes. Casually and unpretentiously leaning against a smooth light gray studio wall; hands are in pockets and one leg is crossed over the other, with relaxed and confident body language. Add to the wall next to them a prominent vector portrait in black and white of their face and bust - with sharp lines and angles, overlapping polygonal shapes and a minimalist modern graphic style, right below the information: “[PERSON’S FULL NAME]”, and below the name: “[PROFESSION]”"
)
puts result.id
package main
import (
"context"
"fmt"
"log"
"net/http"
"os"
"strings"
)
func main() {
body := strings.NewReader("{\"model\":\"nano-banana-2\",\"prompt\":\"Ultra-realistic 8K full body portrait of [PERSON’S FULL NAME], wearing a clean and pressed white social shirt with folded collar and a small lapel microphone, dark navy-blue dress pants and polished brown social shoes. Casually and unpretentiously leaning against a smooth light gray studio wall; hands are in pockets and one leg is crossed over the other, with relaxed and confident body language. Add to the wall next to them a prominent vector portrait in black and white of their face and bust - with sharp lines and angles, overlapping polygonal shapes and a minimalist modern graphic style, right below the information: “[PERSON’S FULL NAME]”, and below the name: “[PROFESSION]”\"}")
req, err := http.NewRequestWithContext(context.Background(), http.MethodPost, "https://runapi.ai/api/v1/nano_banana/text_to_image", body)
if err != nil {
log.Fatal(err)
}
req.Header.Set("Authorization", "Bearer "+os.Getenv("RUNAPI_API_KEY"))
req.Header.Set("Content-Type", "application/json")
resp, err := http.DefaultClient.Do(req)
if err != nil {
log.Fatal(err)
}
defer resp.Body.Close()
fmt.Println(resp.Status)
}
nano-banana-2/api/v1/nano_banana/text_to_imageGet API Key
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "3D avatar of a man, with a happy face on a white background, no necks. Conceptual playlist style digital art with Pixar quality aesthetics"
}
JSON
Ultra-realistic product photography of a chilled blue Pepsi can placed upright on fresh green grass, covered with cold water droplets and condensation. Ice cubes scattered around the base. The can features a vibrant, colorful illustration of Mario and Luigi standing together in a whimsical mushroom kingdom scene with red and blue mushrooms, green pipes, question blocks, and a castle in the background. Bright, cheerful, cartoon-style artwork printed on the can. Shot outdoors in natural daylight with a soft, blurred garden background (bokeh effect), bright green foliage and sunlight filtering through trees. Shallow depth of field, macro detail, high contrast, vivid colors, sharp focus on the can, cinematic product lighting, summer refreshment mood, commercial advertising style, ultra-detailed, 8K quality, photorealistic, centered composition, vertical framing.
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "Ultra-realistic product photography of a chilled blue Pepsi can placed upright on fresh green grass, covered with cold water droplets and condensation. Ice cubes scattered around the base. The can features a vibrant, colorful illustration of Mario and Luigi standing together in a whimsical mushroom kingdom scene with red and blue mushrooms, green pipes, question blocks, and a castle in the background. Bright, cheerful, cartoon-style artwork printed on the can. Shot outdoors in natural daylight with a soft, blurred garden background (bokeh effect), bright green foliage and sunlight filtering through trees. Shallow depth of field, macro detail, high contrast, vivid colors, sharp focus on the can, cinematic product lighting, summer refreshment mood, commercial advertising style, ultra-detailed, 8K quality, photorealistic, centered composition, vertical framing."
}
JSON
A cute 3D chibi-style character of the uploaded image leaning against a parked mini chibi scooter, wearing a clean monochrome outfit with a t-shirt, straight-fit pants, and minimalist sneakers, holding a takeaway juice cup with a relaxed confident smile, big head and expressive eyes, modern fade haircut, set in a cozy café courtyard with wooden chairs, warm golden sunlight, soft depth of field, pastel color palette, smooth rounded proportions, subtle texture detailing, global illumination with gentle rim light, soft natural shadows, playful yet premium Pixar-quality 3D render, toy-like finish, Google Gemini cinematic but cozy mood, ultra-clean high-detail output, Instagram-ready composition, square aspect ratio 1:1
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "A cute 3D chibi-style character of the uploaded image leaning against a parked mini chibi scooter, wearing a clean monochrome outfit with a t-shirt, straight-fit pants, and minimalist sneakers, holding a takeaway juice cup with a relaxed confident smile, big head and expressive eyes, modern fade haircut, set in a cozy café courtyard with wooden chairs, warm golden sunlight, soft depth of field, pastel color palette, smooth rounded proportions, subtle texture detailing, global illumination with gentle rim light, soft natural shadows, playful yet premium Pixar-quality 3D render, toy-like finish, Google Gemini cinematic but cozy mood, ultra-clean high-detail output, Instagram-ready composition, square aspect ratio 1:1"
}
JSON
Transform the referenced person into a premium stylized 3D animated character, full body hero shot, standing confidently in the exact center of the frame. Preserve the exact identity and key facial details as closely as possible, including face shape, eyes, nose, mouth, hairstyle, skin tone, and expression, without redesigning the face. Use soft rounded geometry, smooth polished materials, subtle glossy shading, and a modern premium 3D animation style. The pose must feel dynamic, charismatic, and heroic, not stiff. Let the arms and hands naturally perform an expressive gesture that best suits the character’s body shape, personality, and visual balance, chosen intelligently for the most flattering and appealing result. Keep the full body fully visible while the character still feels close, clear, and strongly present to the camera. Use a clean solid background or a soft gradient [BACKGROUND COLOR]. Studio-quality lighting, soft grounded shadow, strong subject separation, clean premium composition, ultra refined, highly detailed, no props, no text, no logos, no watermark
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "Transform the referenced person into a premium stylized 3D animated character, full body hero shot, standing confidently in the exact center of the frame. Preserve the exact identity and key facial details as closely as possible, including face shape, eyes, nose, mouth, hairstyle, skin tone, and expression, without redesigning the face. Use soft rounded geometry, smooth polished materials, subtle glossy shading, and a modern premium 3D animation style. The pose must feel dynamic, charismatic, and heroic, not stiff. Let the arms and hands naturally perform an expressive gesture that best suits the character’s body shape, personality, and visual balance, chosen intelligently for the most flattering and appealing result. Keep the full body fully visible while the character still feels close, clear, and strongly present to the camera. Use a clean solid background or a soft gradient [BACKGROUND COLOR]. Studio-quality lighting, soft grounded shadow, strong subject separation, clean premium composition, ultra refined, highly detailed, no props, no text, no logos, no watermark"
}
JSON
High-detail 8K split-style composition where the subject is divided vertically into two halves. LEFT SIDE: Keep the original photo fully realistic. Preserve natural lighting, textures, and fine details. Slightly enhance sharpness and clarity. Add subtle hand-drawn doodle outlines (soft pastel or neon lines) around the subject for a creative aura. RIGHT SIDE: Transform the same subject into a stylized cartoon/digital illustration. Maintain the exact same pose, angle, and facial expression. Use clean lineart, soft shading, and pastel or vibrant colors in a modern cute style (chibi / semi-realistic). Add small decorative elements like sticker-style doodles, cute icons, or mini characters floating around. BACKGROUND (IMPORTANT): Keep the ORIGINAL background unchanged on BOTH sides. Do not replace, redraw, or stylize the background. Maintain consistency across the entire image. CENTER DIVISION: Add a vertical torn-paper or glitch-style transition separating the realistic and illustrated halves. STYLE: modern digital art, clean composition, soft lighting, aesthetically pleasing, social media style QUALITY: 8K resolution, ultra-detailed.
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "High-detail 8K split-style composition where the subject is divided vertically into two halves. LEFT SIDE: Keep the original photo fully realistic. Preserve natural lighting, textures, and fine details. Slightly enhance sharpness and clarity. Add subtle hand-drawn doodle outlines (soft pastel or neon lines) around the subject for a creative aura. RIGHT SIDE: Transform the same subject into a stylized cartoon/digital illustration. Maintain the exact same pose, angle, and facial expression. Use clean lineart, soft shading, and pastel or vibrant colors in a modern cute style (chibi / semi-realistic). Add small decorative elements like sticker-style doodles, cute icons, or mini characters floating around. BACKGROUND (IMPORTANT): Keep the ORIGINAL background unchanged on BOTH sides. Do not replace, redraw, or stylize the background. Maintain consistency across the entire image. CENTER DIVISION: Add a vertical torn-paper or glitch-style transition separating the realistic and illustrated halves. STYLE: modern digital art, clean composition, soft lighting, aesthetically pleasing, social media style QUALITY: 8K resolution, ultra-detailed."
}
JSON
A premium 3D stylized version of attached logo, rendered as thick, translucent crystal glass with high refraction. The object is floating amidst fluffy, realistic white cumulus clouds against a bright, clear blue sky. Soft natural sunlight creates bright caustic reflections and rainbow dispersion within the glass. Ethereal, airy, and highly polished 3D aesthetic
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "A premium 3D stylized version of attached logo, rendered as thick, translucent crystal glass with high refraction. The object is floating amidst fluffy, realistic white cumulus clouds against a bright, clear blue sky. Soft natural sunlight creates bright caustic reflections and rainbow dispersion within the glass. Ethereal, airy, and highly polished 3D aesthetic"
}
JSON
Cute stylized 3D animated boy with messy fluffy blue hair and big expressive eyes, soft freckles on face, wearing a fuzzy blue striped sweater, small adorable raccoon sitting on his shoulder, looking upward with a gentle smile, Pixar-style character design, soft studio lighting, pastel blue background, ultra-detailed fur and fabric texture, cinematic portrait composition, high-quality 3D render, 8K.
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "Cute stylized 3D animated boy with messy fluffy blue hair and big expressive eyes, soft freckles on face, wearing a fuzzy blue striped sweater, small adorable raccoon sitting on his shoulder, looking upward with a gentle smile, Pixar-style character design, soft studio lighting, pastel blue background, ultra-detailed fur and fabric texture, cinematic portrait composition, high-quality 3D render, 8K."
}
JSON
A stylized 3D cartoon character of a young boy with big expressive eyes and a slightly sad, pouty expression, standing against a solid pastel green background. He is wearing a soft white fuzzy beanie with ear flaps and matching white over-ear headphones. His hair is short, brown, and slightly messy, peeking out from under the cap. He wears an oversized light green t-shirt with a bold black graphic on the front, and his arms are crossed in a bored or annoyed pose. The lighting is soft, diffused studio lighting with smooth shadows, giving a clean aesthetic. Ultra-detailed, high-quality 3D render, Pixar-style, soft textures, shallow depth of field, centered composition, minimal background, vibrant but soft color palette. Negative prompt (optional): blurry, low quality, extra limbs, distorted face, harsh shadows, cluttered background, text artifacts, watermark
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "A stylized 3D cartoon character of a young boy with big expressive eyes and a slightly sad, pouty expression, standing against a solid pastel green background. He is wearing a soft white fuzzy beanie with ear flaps and matching white over-ear headphones. His hair is short, brown, and slightly messy, peeking out from under the cap. He wears an oversized light green t-shirt with a bold black graphic on the front, and his arms are crossed in a bored or annoyed pose. The lighting is soft, diffused studio lighting with smooth shadows, giving a clean aesthetic. Ultra-detailed, high-quality 3D render, Pixar-style, soft textures, shallow depth of field, centered composition, minimal background, vibrant but soft color palette. Negative prompt (optional): blurry, low quality, extra limbs, distorted face, harsh shadows, cluttered background, text artifacts, watermark"
}
JSON
A hyper-realistic product photo of a Kinder Joy surprise egg sitting on a wooden table. The egg packaging features a chibi/cartoon-style illustration of [image Uploaded] posed against a graffiti brick wall background. Next to the egg sits a small detailed collectible figurine of [image Uploaded] in the same chibi art style shown on the packaging. Soft natural lighting, shallow depth of field, photorealistic quality.
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "A hyper-realistic product photo of a Kinder Joy surprise egg sitting on a wooden table. The egg packaging features a chibi/cartoon-style illustration of [image Uploaded] posed against a graffiti brick wall background. Next to the egg sits a small detailed collectible figurine of [image Uploaded] in the same chibi art style shown on the packaging. Soft natural lighting, shallow depth of field, photorealistic quality."
}
JSON
A detailed, collectible-style 3D render of a linear procession of twelve identical figures, viewed in a dynamic diagonal 3/4 perspective down the line. The figures are arranged on a dark, gradient studio floor with a progressively shallower depth of field, making the furthest characters less detailed. Specific Pose Sequence: The procession must follow a distinct sequence of poses along the line: The first and closest figure stands with arms at their sides. The second figure stands with arms crossed over their chest. The third figure stands with one hand raised in a salute or wave. The subsequent receding figures doing different poses. Lighting and Atmosphere: The scene features dramatic, atmospheric backlighting from a powerful source behind the furthest characters, creating a bright halo and deep rim-lighting, combined with a strong, texturing haze or fog. Dramatic, focused front-side lighting illuminates the features of the nearest character. The background fades to complete shadow. Fixed Elements: The overall style is a high-end, textured, 3D vinyl toy aesthetic. Camera: Low-to-medium camera angle looking down the diagonal line. Camera & Framing Refinement (CRITICAL): Shot with a close-range 28–35mm lens to introduce natural perspective compression without distortion. The camera is positioned slightly below chest level of the first figure and very close to it, creating strong foreground dominance. The first figure occupies approximately 65–75% of the frame height and must be tightly cropped at knee level. The feet of the first figure must NOT be visible. The head of the first figure sits very close to the top edge of the frame with minimal breathing room. The figure should feel slightly too large for the frame, as if it cannot fully fit inside. Subsequent figures progressively reveal more of their full bodies as they recede into depth. The diagonal line of figures must start large in the foreground (left side of frame) and recede tightly into the background (right side), with each figure overlapping slightly with the next. The camera is not centered to the line. It is offset to the left, looking across the figures, not directly along them. Spacing between figures compresses progressively toward the background, enhancing depth and scale illusion. No wide empty gaps between characters. The line must feel dense and continuous. 2D to 3D Translation (ONLY if input is 2D artwork): If the uploaded reference is 2D (illustration, anime, flat graphic, cel-shaded), convert it into a fully realized 3D cinematic form while preserving exact identity fidelity. • Maintain exact head shape, eye design, spacing, and proportions • Preserve original stylization, do NOT add realism that alters design language • Translate linework into subtle surface geometry or shading transitions • Keep original colors, expressed through physically believable materials • Convert flat shading into controlled cinematic light falloff The result must feel physically real, not redesigned. ar 1:1
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "A detailed, collectible-style 3D render of a linear procession of twelve identical figures, viewed in a dynamic diagonal 3/4 perspective down the line. The figures are arranged on a dark, gradient studio floor with a progressively shallower depth of field, making the furthest characters less detailed. Specific Pose Sequence: The procession must follow a distinct sequence of poses along the line: The first and closest figure stands with arms at their sides. The second figure stands with arms crossed over their chest. The third figure stands with one hand raised in a salute or wave. The subsequent receding figures doing different poses. Lighting and Atmosphere: The scene features dramatic, atmospheric backlighting from a powerful source behind the furthest characters, creating a bright halo and deep rim-lighting, combined with a strong, texturing haze or fog. Dramatic, focused front-side lighting illuminates the features of the nearest character. The background fades to complete shadow. Fixed Elements: The overall style is a high-end, textured, 3D vinyl toy aesthetic. Camera: Low-to-medium camera angle looking down the diagonal line. Camera & Framing Refinement (CRITICAL): Shot with a close-range 28–35mm lens to introduce natural perspective compression without distortion. The camera is positioned slightly below chest level of the first figure and very close to it, creating strong foreground dominance. The first figure occupies approximately 65–75% of the frame height and must be tightly cropped at knee level. The feet of the first figure must NOT be visible. The head of the first figure sits very close to the top edge of the frame with minimal breathing room. The figure should feel slightly too large for the frame, as if it cannot fully fit inside. Subsequent figures progressively reveal more of their full bodies as they recede into depth. The diagonal line of figures must start large in the foreground (left side of frame) and recede tightly into the background (right side), with each figure overlapping slightly with the next. The camera is not centered to the line. It is offset to the left, looking across the figures, not directly along them. Spacing between figures compresses progressively toward the background, enhancing depth and scale illusion. No wide empty gaps between characters. The line must feel dense and continuous. 2D to 3D Translation (ONLY if input is 2D artwork): If the uploaded reference is 2D (illustration, anime, flat graphic, cel-shaded), convert it into a fully realized 3D cinematic form while preserving exact identity fidelity. • Maintain exact head shape, eye design, spacing, and proportions • Preserve original stylization, do NOT add realism that alters design language • Translate linework into subtle surface geometry or shading transitions • Keep original colors, expressed through physically believable materials • Convert flat shading into controlled cinematic light falloff The result must feel physically real, not redesigned. ar 1:1"
}
JSON
2x2 grid, do this for 4 important years in history and their most important inventions { Anchor: `[Letter/Number] :: [Era/Theme (e.g., 1980s Retro Tech / Victorian Steampunk)]::4` Morphology: `3D Typographic sculpture of [Letter/Number], character shape defined by a dense pile of [Era/Theme] objects, chaotic but legible assemblage, graphic design asset::3` Material Physics: `Textures of [Era/Theme] objects (plastic, rust, wood), occlusion between stacked items, realistic scale of constituent parts forming the glyph::3` Illumination: `Three-point studio lighting, softbox overhead, clean background for isolation::2` Render Stack: `Cinema 4D render, isometric or front-facing, ambient occlusion, 8k, Adobe Substance material suite::1` Negative: `[Illegible, flat font, 2D, handwriting, paper, messy background, floating objects]:: -1`}
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "2x2 grid, do this for 4 important years in history and their most important inventions { Anchor: `[Letter/Number] :: [Era/Theme (e.g., 1980s Retro Tech / Victorian Steampunk)]::4` Morphology: `3D Typographic sculpture of [Letter/Number], character shape defined by a dense pile of [Era/Theme] objects, chaotic but legible assemblage, graphic design asset::3` Material Physics: `Textures of [Era/Theme] objects (plastic, rust, wood), occlusion between stacked items, realistic scale of constituent parts forming the glyph::3` Illumination: `Three-point studio lighting, softbox overhead, clean background for isolation::2` Render Stack: `Cinema 4D render, isometric or front-facing, ambient occlusion, 8k, Adobe Substance material suite::1` Negative: `[Illegible, flat font, 2D, handwriting, paper, messy background, floating objects]:: -1`}"
}
JSON
A stunning Trompe L’oeil 3D composition, where the subject—using an uploaded face as reference—appears to physically tear through a clean white drawing paper and emerge from a graphite sketch into the real world. The subject’s upper body is fully rendered in hyper-realistic color and texture, wearing the same clothes from the uploaded image, while the lower body remains a 2D pencil sketch on paper. Shot from an aerial (top-down) perspective using a 24mm lens to capture the entire drawing desk environment, including oversized pencils and erasers scattered around. The lighting is a blend of flat studio light for the paper area and dramatic side lighting with volumetric effects for the emerging 3D figure, casting realistic shadows that fall back onto the sketch area. 8K resolution, ray tracing, and advanced shading techniques are used to blur the boundary between illustration and reality in this extraordinary surrealist piece.
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "A stunning Trompe L’oeil 3D composition, where the subject—using an uploaded face as reference—appears to physically tear through a clean white drawing paper and emerge from a graphite sketch into the real world. The subject’s upper body is fully rendered in hyper-realistic color and texture, wearing the same clothes from the uploaded image, while the lower body remains a 2D pencil sketch on paper. Shot from an aerial (top-down) perspective using a 24mm lens to capture the entire drawing desk environment, including oversized pencils and erasers scattered around. The lighting is a blend of flat studio light for the paper area and dramatic side lighting with volumetric effects for the emerging 3D figure, casting realistic shadows that fall back onto the sketch area. 8K resolution, ray tracing, and advanced shading techniques are used to blur the boundary between illustration and reality in this extraordinary surrealist piece."
}
JSON
1080x1080 cork-board layout telling the story of [HISTORICAL MOMENT]. At the very top of the board, include a pinned paper label with the title: "HISTORICAL MOMENT" in large handwritten letters. Below it, arrange 5–6 Polaroid photos pinned across the board in a loose chronological path. Each Polaroid shows a key moment from the story with a short handwritten caption beneath it. Connect characters and events with colored strings (red, blue, yellow). Warm nostalgic lighting, soft shadows. Include realistic details: coffee cup rings, paper clips, torn notes, arrows, scribbled hints, and thumbtacks. Vintage, textured, cozy detective-board aesthetic. Highly readable, high contrast, everything framed clearly for 1080x1080
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "1080x1080 cork-board layout telling the story of [HISTORICAL MOMENT]. At the very top of the board, include a pinned paper label with the title: \"HISTORICAL MOMENT\" in large handwritten letters. Below it, arrange 5–6 Polaroid photos pinned across the board in a loose chronological path. Each Polaroid shows a key moment from the story with a short handwritten caption beneath it. Connect characters and events with colored strings (red, blue, yellow). Warm nostalgic lighting, soft shadows. Include realistic details: coffee cup rings, paper clips, torn notes, arrows, scribbled hints, and thumbtacks. Vintage, textured, cozy detective-board aesthetic. Highly readable, high contrast, everything framed clearly for 1080x1080"
}
JSON
8K Ultra-Realistic Promotional Design a hyper-realistic close-up image of a folded sheet of aged paper. Within the paper’s creases and surfaces, depict a [SCENE OR MEMORY] emerging naturally from the folds, like sculpted origami. The environment and figures (if present) appear embossed or printed directly into the paper, using wrinkles, grain, and ink textures to build the illusion. Include subtle shadows, curled edges, torn corners, and ink fading to enhance realism. Set the paper on a neutral tabletop with soft directional light from one side, casting gentle shadows and enhancing dimensional depth. Photorealistic, emotional storytelling, handcrafted visual illusion, 8K ultra-detail, and image size 4 5.
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "8K Ultra-Realistic Promotional Design a hyper-realistic close-up image of a folded sheet of aged paper. Within the paper’s creases and surfaces, depict a [SCENE OR MEMORY] emerging naturally from the folds, like sculpted origami. The environment and figures (if present) appear embossed or printed directly into the paper, using wrinkles, grain, and ink textures to build the illusion. Include subtle shadows, curled edges, torn corners, and ink fading to enhance realism. Set the paper on a neutral tabletop with soft directional light from one side, casting gentle shadows and enhancing dimensional depth. Photorealistic, emotional storytelling, handcrafted visual illusion, 8K ultra-detail, and image size 4 5."
}
JSON
Ultra-detailed large-scale miniature diorama of [STADIUM NAME] in [CITY] presented as a premium museum-quality desktop exhibition model. True architectural scale model craftsmanship with extreme fine detail. [STADIUM NAME] fully visible and centered in the composition: iconic roof structure, exterior facade details, floodlight towers rising dramatically, the entire stadium accurately represented from ground level to roof. Surrounding miniature [CITY] city blocks: small local-style residential buildings, grid street layout, tiny trees lining the sidewalks, mini cars and pedestrians for scale realism. Ground textures: stone pavements, micro streetlights, tiny fan plazas and entrance gates surrounding the stadium. Materials: realistic miniature concrete and steel texture, matte architectural model surfaces, precise structural detailing, museum-grade craftsmanship. Display base: elegant rectangular exhibition base, subtle engraved metal plate reading “[CITY] – [STADIUM NAME]”. Lighting: soft museum spotlighting from above, gentle shadows enhancing the structural details, clean studio environment. Camera: slightly elevated macro architectural photography angle, tilt-shift miniature aesthetic, center extremely sharp with slight edge depth falloff. Ultra realistic collectible stadium model, 8K resolution, luxury museum exhibition presentation. Aspect ratio 4:5 vertical
Ultra-realistic studio portrait of an anthropomorphic rooster styled like a street skater. A confident white rooster with a red comb and wattles stands upright like a human, full body visible, facing slightly to the side. The rooster is wearing oversized vintage blue denim jeans with a loose, baggy fit, a faded denim jacket hanging off the shoulders, and a thick silver chain necklace around its neck. A brown leather belt is fastened tightly around the waist. The rooster is holding a worn skateboard vertically with one wing, resting it on the floor. The skateboard has scratched trucks, beige wheels, and a distressed, colorful deck. On its feet are chunky white high-top sneakers with visible wear and realistic fabric texture. Highly detailed feathers with realistic texture and volume, sharp focus on the rooster’s face and clothing. Studio lighting with soft shadows, clean white/neutral background, fashion editorial style. Cinematic realism, humorous yet stylish vibe, perfect balance between realism and surreal character design. Ultra-high resolution, hyper-detailed materials (denim fabric, leather belt, metal chain, feathers), shallow depth of field, professional photography look.
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "Ultra-realistic studio portrait of an anthropomorphic rooster styled like a street skater. A confident white rooster with a red comb and wattles stands upright like a human, full body visible, facing slightly to the side. The rooster is wearing oversized vintage blue denim jeans with a loose, baggy fit, a faded denim jacket hanging off the shoulders, and a thick silver chain necklace around its neck. A brown leather belt is fastened tightly around the waist. The rooster is holding a worn skateboard vertically with one wing, resting it on the floor. The skateboard has scratched trucks, beige wheels, and a distressed, colorful deck. On its feet are chunky white high-top sneakers with visible wear and realistic fabric texture. Highly detailed feathers with realistic texture and volume, sharp focus on the rooster’s face and clothing. Studio lighting with soft shadows, clean white/neutral background, fashion editorial style. Cinematic realism, humorous yet stylish vibe, perfect balance between realism and surreal character design. Ultra-high resolution, hyper-detailed materials (denim fabric, leather belt, metal chain, feathers), shallow depth of field, professional photography look."
}
JSON
A high-speed studio photograph of a cluster of fresh blueberries suspended midair, softly wrapped by a gentle flow of pale blueberry yogurt, smooth and restrained liquid motion, minimal splash, thick and creamy texture, elegant curved yogurt flow hugging the berries, very few small droplets, calm fluid dynamics, photorealistic macro food photography, ultra-realistic texture, professional studio lighting, soft diffuse highlights, pure white background, isolated object, clean cutout, no CGI, no 3D render, no illustration
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "A high-speed studio photograph of a cluster of fresh blueberries suspended midair, softly wrapped by a gentle flow of pale blueberry yogurt, smooth and restrained liquid motion, minimal splash, thick and creamy texture, elegant curved yogurt flow hugging the berries, very few small droplets, calm fluid dynamics, photorealistic macro food photography, ultra-realistic texture, professional studio lighting, soft diffuse highlights, pure white background, isolated object, clean cutout, no CGI, no 3D render, no illustration"
}
JSON
Cinematic 8k hyper-realistic close-up shot. A 3D realistic Bugs Bunny, faithful to his classic design but with detailed fur texture, wearing a white barber uniform. He is holding a bottle labeled "HARE TONIC" and pouring liquid onto the head of a sitting man. Man [photos reference with hair] has a worried and nervous expression, water dripping down his face. Vintage barbershop background with a mirror. Dynamic lighting, high detail, photorealistic rendering style
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "Cinematic 8k hyper-realistic close-up shot. A 3D realistic Bugs Bunny, faithful to his classic design but with detailed fur texture, wearing a white barber uniform. He is holding a bottle labeled \"HARE TONIC\" and pouring liquid onto the head of a sitting man. Man [photos reference with hair] has a worried and nervous expression, water dripping down his face. Vintage barbershop background with a mirror. Dynamic lighting, high detail, photorealistic rendering style"
}
JSON
Input Variable: [Any Famous Building - Guggenheim, Burj Khalifa, Sydney Opera House, etc.] System Instruction: Generate a hyper-realistic, isometric 3D "Exploded View" diorama showing the Input Building deconstructed into its structural systems with an architectural model-making crew assembling a scale version. 1. Semantic Engineering Analysis: Layer 1 (The Skin): Exterior facade panels, windows, cladding, roofing, entrance canopy Layer 2 (The Structure): Steel framework, floor plates, columns, beams, elevator shafts, stairwells, mechanical systems Layer 3 (The Foundation): Basement levels, support pillars, utility conduits, HVAC units, individual rivets/bolts, rebar cages 2. Container: Surface: Architect's drafting table with cutting mats and T-squares Layout: Building layers stacked vertically showing ground-up construction sequence Connections: White architectural section lines showing structural load paths and circulation Additional Elements: Blueprint sheets, scale rulers, material samples, elevation drawings 3. Micro-Narrative: CRITICAL: This is meta - tiny workers building a model that's being built by tiny workers. Recursive scale concept. Workers: 1:87 scale model makers in black turtlenecks (architect stereotype) Equipment: Laser cutters, 3D printers, basswood cutting stations, acrylic gluing rigs, miniature cranes positioning floor plates Interaction: Architects cutting foam core, assembling balsa wood frames, installing tiny windows, spray painting facades, arranging scale furniture 4. Visual Syntax & Material Physics: Materials: Basswood grain, clear acrylic sheets, white foam core, brass rod framework, museum board textures Lighting: Architecture studio track lighting creating dramatic model shadows Special Effects: Glue strands, wood shavings, spray paint mist, laser cutter burn marks Output: Elevated three-quarter view, architectural presentation, 1:1 ratio, 8K, Archdaily model photography aesthetic
Generate a 3D isometric diorama illustration of [CHARACTER] working from home at a desk with a laptop and a window. Handmade felt appearance with rounded polished shapes, enclosed in a hexagonal rounded cube frame. The room decor, colours, furniture, wall art, and accessories are all instantly recognisable to [CHARACTER] — reflecting their signature style, iconic items, outfit, and personality. Warm soft lighting, clean white background, 1080x1080, photorealistic quality. No text labels or signs on the base.
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "Generate a 3D isometric diorama illustration of [CHARACTER] working from home at a desk with a laptop and a window. Handmade felt appearance with rounded polished shapes, enclosed in a hexagonal rounded cube frame. The room decor, colours, furniture, wall art, and accessories are all instantly recognisable to [CHARACTER] — reflecting their signature style, iconic items, outfit, and personality. Warm soft lighting, clean white background, 1080x1080, photorealistic quality. No text labels or signs on the base."
}
JSON
Use the image I sent. Create a hand-drawn portrait illustration with red and yellow pen on notebook paper, inspired by comic book drawings and annotations. Maintain a complete likeness to the subject, expressive lines, spontaneous gestures, sharp and bright outlines, handwritten annotations around the edges, and a realistic texture of the pen strokes. AR 3:4
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "Use the image I sent. Create a hand-drawn portrait illustration with red and yellow pen on notebook paper, inspired by comic book drawings and annotations. Maintain a complete likeness to the subject, expressive lines, spontaneous gestures, sharp and bright outlines, handwritten annotations around the edges, and a realistic texture of the pen strokes. AR 3:4"
}
JSON
3D Pixar-style animated child character, cute young boy with messy brown hair and big expressive blue eyes, surprised open-mouth expression, soft round face, wearing a fuzzy teal sweater with small “noo” text on chest, loose beige patterned pants, white sneakers, small brown crossbody bag, standing on an orange electric scooter, minimal teal studio background, soft lighting, ultra-detailed textures (fabric fuzz, skin, hair), cinematic depth of field, vibrant colors, highly polished render, octane render, 4k
curl -X POST https://runapi.ai/api/v1/nano_banana/text_to_image \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'JSON'
{
"model": "nano-banana-2",
"prompt": "3D Pixar-style animated child character, cute young boy with messy brown hair and big expressive blue eyes, surprised open-mouth expression, soft round face, wearing a fuzzy teal sweater with small “noo” text on chest, loose beige patterned pants, white sneakers, small brown crossbody bag, standing on an orange electric scooter, minimal teal studio background, soft lighting, ultra-detailed textures (fabric fuzz, skin, hair), cinematic depth of field, vibrant colors, highly polished render, octane render, 4k"
}
JSON
FAQ
Using nano-banana-2 prompts
What is %{model}?
%{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.