About this tool

The IteraTools Image Description API uses GPT-4o vision to produce detailed, contextual descriptions of any image. Pass a public image URL or base64-encoded image data, and receive a rich natural-language description plus semantic tags. Falls back to Claude 3 Haiku automatically if GPT-4o is unavailable. Perfect for AI agents that need to reason about visual content, build alt-text pipelines, classify images, or understand user-uploaded media.

🧪 Try it live

API Key

Image URL

Custom prompt (optional)

Quick Start

curl -X POST https://api.iteratools.com/image/describe \
  -H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "image": "https://example.com/photo.jpg",
    "prompt": "What objects and people are in this image?"
  }'

Response

{
  "ok": true,
  "data": {
    "description": "The image shows a wooden desk with a laptop computer open to a code editor...",
    "tags": ["laptop", "desk", "code", "monitor", "keyboard", "office", "workspace"],
    "model": "gpt-4o"
  }
}

Parameters

Param	Type	Required	Description
image	string	Yes	Public HTTPS URL or base64 data URL (`data:image/jpeg;base64,...`). Max 10MB.
prompt	string	No	Custom question or instruction. Default: "Describe this image in detail. List objects, people, text, colors, and context."

Supported formats

JPEG PNG GIF WEBP Max 10MB

Use cases

Generate alt-text for accessibility compliance
Classify or tag user-uploaded images automatically
Extract product details from e-commerce photos
Understand screenshots in customer support workflows
Feed visual context into AI agent reasoning pipelines
Describe diagrams, charts, and technical illustrations
Moderate or categorize image content at scale

MCP (Claude / Cursor)

npx -y @iteratools/mcp --key YOUR_API_KEY

Adds all 51 IteraTools tools to your MCP client — including image_describe.

Pricing

$0.008 per request. Pay with API credits (no subscription). Get your API key →