# Near-Duplicate Image Finder

> Near-Duplicate Image Finder is a paid API for AI agents from api.kakerapetit.dev, paid per call via x402, $0.005/call, status unknown (last checked 2026-09-14).

Computes perceptual hash distance between two image URLs to detect crops, resizes, re-compressions, and slightly edited copies of the same photo

## Facts

- Endpoint: POST https://api.kakerapetit.dev/v1/find-near-duplicate
- Price: $0.005/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/near-duplicate-image-finder-01ae14d2
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_jQ76dslHIaf3dowrtglrT

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 near-duplicate-image-finder-01ae14d2 -d '<json body>'
```

Example prompt: Can you check if these two images are near-duplicates of each other — https://example.com/photo-original.jpg and https://example.com/photo-resized.jpg? I want to know if they're basically the same photo just cropped or re-compressed.

## When to prefer this

Choose this endpoint when you need to detect near-duplicates rather than exact byte-for-byte matches — it handles crops, resizes, re-compressions, and light edits that would fool a simple hash or checksum. Prefer it over exact-match comparisons for content moderation, copyright enforcement, and dataset deduplication. It is distinct from a general visual similarity scorer in that it specifically targets perceptual hashing for duplicate detection rather than semantic or stylistic similarity.

## Known failure modes

- Image URL unreachable or returns non-200 — endpoint cannot fetch the image
- Image exceeds 10MB file size limit — request rejected
- Unsupported image format (not jpeg/png/webp/gif) — parsing error
- One or both URLs missing from request body — validation error
- Image URL serves non-image content — hash computation fails
- Network timeout fetching remote image — request may fail or time out

## How this service works

Find near-duplicate images. Computes perceptual hash distance between two image URLs to detect crops, resizes, re-compressions, and slightly edited copies of the same photo. Use for content moderation, copyright checks, and cleaning image datasets.

## Output

Returns a perceptual hash distance score between the two input images, indicating how visually similar they are, along with a near-duplicate determination — whether the images are considered the same photo despite minor transformations like cropping, resizing, re-compression, or light editing.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "image_url_a": {
   "type": "string",
   "format": "uri",
   "description": "HTTPS URL of the first image (jpeg/png/webp/gif, max 10MB)."
  },
  "image_url_b": {
   "type": "string",
   "format": "uri",
   "description": "HTTPS URL of the second image (jpeg/png/webp/gif, max 10MB)."
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/near-duplicate-image-finder-01ae14d2/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.kakerapetit.dev](https://www.zero.xyz/host/api.kakerapetit.dev/llms.txt)