# Diffbot Image Extraction

> Diffbot Image Extraction is a paid API for AI agents from Diffbot at diffbot.mpp.paywithlocus.com, paid per call via MPP, $0.0042/call, status unknown (last checked 2026-10-12).

Extracts structured metadata from an image page URL using Diffbot's AI-powered web extraction

## Facts

- Endpoint: POST https://diffbot.mpp.paywithlocus.com/diffbot/image?utm_source=zero.xyz
- Price: $0.0042/call
- Payment: MPP
- Status: unknown
- Last checked: 2026-10-12
- Activations on Zero: 0
- Provider: Diffbot
- Docs: https://docs.diffbot.com
- Website: https://www.diffbot.com
- Tags: web-scraping, extraction, articles, products, web, data
- Canonical page: https://www.zero.xyz/c/diffbot-image-c930e5cf
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_9iZFpZwbMUwYow3h3EaT0

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability diffbot-image-c930e5cf -d '<json body>'
```

Example prompt: Use Diffbot to extract structured image metadata from this page: https://www.example.com/photo/sunset-over-mountains — I want the title, caption, author, tags, and any links on the page.

## When to prefer this

Use this endpoint when you need to extract structured metadata specifically from image pages on the web, leveraging Diffbot's AI-powered extraction rather than writing custom scraping logic. Prefer this over generic HTML scrapers when you want clean, typed fields like title, caption, author, and tags without manual parsing.

## Known failure modes

- URL is not an image page — returns empty or misclassified data
- Target page is behind a login or paywall — extraction returns partial or no data
- Timeout exceeded for slow-loading pages — returns timeout error
- URL is malformed or unreachable — returns 4xx or connection error
- Page has anti-scraping measures — returns empty or blocked response

## How this service works

Extract image details from pages — URLs, dimensions, titles, alt text.

## Output

Returns structured JSON with extracted image metadata including image URL, title, caption, alt text, author, source, dimensions, tags, and optionally links, breadcrumbs, and discussion threads found on the page.

## Example request

```json
{
 "url": "https://en.wikipedia.org/wiki/File:Camponotus_flavomarginatus_ant.jpg",
 "fields": "links,meta,breadcrumb",
 "timeout": 30000
}
```

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "url"
 ],
 "properties": {
  "url": {
   "type": "string",
   "description": "URL of the image page"
  },
  "fields": {
   "type": "string",
   "description": "Comma-separated extra fields to return (e.g. links, meta, breadcrumb)"
  },
  "timeout": {
   "type": "number",
   "description": "Request timeout in ms (default 30000)"
  },
  "discussion": {
   "type": "boolean",
   "description": "Extract comments/discussion threads (default true for article)"
  }
 },
 "additionalProperties": true
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/diffbot-image-c930e5cf/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from diffbot.mpp.paywithlocus.com](https://www.zero.xyz/host/diffbot.mpp.paywithlocus.com/llms.txt)