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.
A 3x3 grid storyboard contact sheet comprised of nine candid, cinematic movie stills showing a continuous sequence of a young man, approximately 20-25 years old. He has tousled brown hair and a youthful, expressive face, clean-shaven. Across all nine panels, he is consistently wearing a brown short-sleeved linen button-down shirt and is seated at a wooden dining table during a family dinner. The panels show varied natural actions: gesturing with his hands while animatedly talking, laughing, eating lasagna from a large dish, listening intently, and smiling. The environment is a warm, cozy home dining room with a wooden hutch filled with china and patterned wallpaper visible in the background. The lighting is soft, warm tungsten evening light creating natural shadows. The shots are a mix of medium angles and close-ups with a shallow depth of field, rendered with a film photography aesthetic and slight grain.
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 3x3 grid storyboard contact sheet comprised of nine candid, cinematic movie stills showing a continuous sequence of a young man, approximately 20-25 years old. He has tousled brown hair and a youthful, expressive face, clean-shaven. Across all nine panels, he is consistently wearing a brown short-sleeved linen button-down shirt and is seated at a wooden dining table during a family dinner. The panels show varied natural actions: gesturing with his hands while animatedly talking, laughing, eating lasagna from a large dish, listening intently, and smiling. The environment is a warm, cozy home dining room with a wooden hutch filled with china and patterned wallpaper visible in the background. The lighting is soft, warm tungsten evening light creating natural shadows. The shots are a mix of medium angles and close-ups with a shallow depth of field, rendered with a film photography aesthetic and slight grain."
}
JSON
Create a hyper-realistic, stylish vertical poster featuring a smartphone lying on a table, transformed into a hockey rink. The screen resembles cracked ice with deep fissures, while tiny hockey players compete in an intense game Use diffused daylight with elegant reflections and refractions to enhance depth. Add realistic details like fingerprints, smudges, and subtle scratches on the phone's surface for authenticity. Craft a dynamic perspective with an intriguing camera angle-slightly tilted for dramatic effect. The image should look like a high-quality DSLR photograph, with sharp focus, rich textures, and cinematic lighting to amplify the surreal yet lifelike atmosphere
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, stylish vertical poster featuring a smartphone lying on a table, transformed into a hockey rink. The screen resembles cracked ice with deep fissures, while tiny hockey players compete in an intense game Use diffused daylight with elegant reflections and refractions to enhance depth. Add realistic details like fingerprints, smudges, and subtle scratches on the phone's surface for authenticity. Craft a dynamic perspective with an intriguing camera angle-slightly tilted for dramatic effect. The image should look like a high-quality DSLR photograph, with sharp focus, rich textures, and cinematic lighting to amplify the surreal yet lifelike atmosphere"
}
JSON
A high-quality, candid street photography shot of a stylish young woman with short wavy brown hair. She is wearing a maroon cardigan, oversized beige denim cargo pants, and a maroon baseball cap. She has large white over-ear headphones around her ears and a black quilted shoulder bag with a small plush dog keychain hanging from it.Same face as in reference
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 high-quality, candid street photography shot of a stylish young woman with short wavy brown hair. She is wearing a maroon cardigan, oversized beige denim cargo pants, and a maroon baseball cap. She has large white over-ear headphones around her ears and a black quilted shoulder bag with a small plush dog keychain hanging from it.Same face as in reference"
}
JSON
Higgsfield just dropped GPT Image 1.5 — I tried the "impossible" test everyone thought would fail: 1⃣ Pure anime illustration only 2⃣ Extract every clothing item while keeping anime style 3⃣ Take those flat 2D clothes 4⃣ Dress a 100% photorealistic girl with them No LoRA. No training. Just one simple prompt. Result: pleats fold like real fabric, cardigan has actual weight, socks compress naturally. GPT Image 1.5 doesn’t just copy styles — it truly understands clothing physics and materials. Anime uniform on a real human. For real. My brain is broken 🤯✨ #Higgsfield #gptimage
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": "Higgsfield just dropped GPT Image 1.5 — I tried the \"impossible\" test everyone thought would fail: 1⃣ Pure anime illustration only 2⃣ Extract every clothing item while keeping anime style 3⃣ Take those flat 2D clothes 4⃣ Dress a 100% photorealistic girl with them No LoRA. No training. Just one simple prompt. Result: pleats fold like real fabric, cardigan has actual weight, socks compress naturally. GPT Image 1.5 doesn’t just copy styles — it truly understands clothing physics and materials. Anime uniform on a real human. For real. My brain is broken 🤯✨ #Higgsfield #gptimage"
}
JSON
FAQ
Using gpt-image-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.