Authentication & paid access
API access is available to Starter and Pro accounts only. Send your API key in the Authorization: Bearer header. Calls spend credits from the same account balance as browser calls, while the web app keeps its own Free plan experience.
Bring your own provider token
The managed endpoint on this page accepts only FIND_IMAGE_PROMPT_API_KEY. If you prefer your own OpenRouter, OpenAI/GPT, or Gemini account, use the open-source DeepSeek Function Calling adapter. It runs locally or server-side, routes create_image_prompt to the selected provider, and never sends a third-party token to Find Image Prompt.
export DEEPSEEK_API_KEY="your_deepseek_api_key"
export OPENROUTER_API_KEY="your_openrouter_api_key"
node plugin/deepseek/src/index.mjs --provider openrouter --message "Write a Midjourney prompt for a rainy Kyoto bookshop at blue hour"The adapter supports --provider find-image-prompt, openrouter, openai, gemini, and auto. DeepSeek Harness (DSH) plugin authors can reuse its create_image_prompt tool contract; a native installable DSH apply() module is not included yet.
POST /api/v1/prompts
Creates a prompt or image description. It accepts JSON and returns result, creditsUsed, and a supportable requestId.
Examples by operation
Set FIND_IMAGE_PROMPT_API_KEY in your shell or server environment. Never expose it in client-side code.
Text to prompt
Turn a creative brief into a reusable visual prompt.
curl https://findimageprompt.com/api/v1/prompts \
-H "Authorization: Bearer $FIND_IMAGE_PROMPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{"operation":"text_to_prompt","text":"A tiny tea shop on the moon","target_model":"GPT Image"}'Image to prompt
Recreate the style, composition, and visual facts of a public reference image.
curl https://findimageprompt.com/api/v1/prompts \
-H "Authorization: Bearer $FIND_IMAGE_PROMPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{"operation":"image_to_prompt","image_url":"https://example.com/reference.jpg","target_model":"Midjourney","prompt_format":"Detailed prompt"}'Describe an image
Extract visible details, objects, art style, or text from an image.
curl https://findimageprompt.com/api/v1/prompts \
-H "Authorization: Bearer $FIND_IMAGE_PROMPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{"operation":"describe_image","image_url":"https://example.com/poster.webp","description_option":"artStyle","language":"English"}'ChatGPT Images
Create an edit or generation prompt tailored to ChatGPT Images.
curl https://findimageprompt.com/api/v1/prompts \
-H "Authorization: Bearer $FIND_IMAGE_PROMPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{"operation":"chatgpt_image_prompt","text":"Turn this into a warm winter editorial photo","image_url":"https://example.com/portrait.jpg"}'Gemini photo prompt
Create a photo-generation prompt tailored to Gemini.
curl https://findimageprompt.com/api/v1/prompts \
-H "Authorization: Bearer $FIND_IMAGE_PROMPT_API_KEY" \
-H "Content-Type: application/json" \
-d '{"operation":"gemini_photo_prompt","text":"A coastal engagement portrait at blue hour","language":"English"}'Request fields
| Field | Required | Notes |
|---|---|---|
operation | Yes | text_to_prompt, image_to_prompt, describe_image, chatgpt_image_prompt, or gemini_photo_prompt |
text | For text operation | Your idea or desired edit. |
image_url | For image operations | Public HTTPS PNG, JPEG, or WebP; maximum 10 MB. |
target_model | No | Nano Banana, GPT Image, Flux, Midjourney, or Stable Diffusion. |
prompt_format | No | General prompt, Detailed prompt, Structured prompt, Graphic design, or JSON. |
language | No | Output language; English is the default. |
Errors & safety
401 means the key is missing, invalid, or revoked. 402 means the account has insufficient credits. 403 means API access requires a paid plan. 429 signals a capacity limit. Unsafe inputs are rejected with 400. Never expose an API key in browser code or public repositories.