# SimHash Near-Duplicate Detection

> SimHash Near-Duplicate Detection is a paid API for AI agents from api.24klabs.ai, paid per call via x402, $0.002/call, status unknown (last checked 2026-09-15).

Computes a 64-bit SimHash fingerprint for text content, or compares two texts returning Hamming distance and a 0–1 similarity score for near-duplicate detection.

## Facts

- Endpoint: POST https://api.24klabs.ai/api/v1/simhash-dedup
- Price: $0.002/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/simhash-near-duplicate-detection-8e7d3026
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_yFaGOwHEk-781F9bWtgUm

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 simhash-near-duplicate-detection-8e7d3026 -d '<json body>'
```

Example prompt: Compare these two product descriptions and tell me their SimHash similarity score and Hamming distance so I can decide if they're near-duplicates: 'Fast wireless charger for iPhone 15, 20W' and 'iPhone 15 wireless charger, 20W fast charging'.

## When to prefer this

Use this endpoint when you need fast, scalable near-duplicate detection for text without requiring a full semantic embedding model — ideal for deduplication pipelines, plagiarism checks, or content fingerprinting where fuzzy matching is needed. Prefer this over exact-hash approaches (MD5/SHA) when you need to detect near-duplicates with minor edits, and prefer it over expensive embedding-based similarity when exact semantic understanding is not required.

## Known failure modes

- Empty or missing text input returns a validation error
- Extremely long texts may be truncated or cause timeout
- Non-text binary content is not supported
- Similarity score of exactly 0 or 1 may be returned for trivially identical or completely unrelated texts
- Payment failure via x402 results in 402 response

## How this service works

Compute a 64-bit SimHash fingerprint for content, or compare two texts and return the Hamming distance and a 0-1 similarity score for near-duplicate detection.

## Output

Returns a 64-bit SimHash fingerprint for a single input, or when comparing two texts returns the Hamming distance (integer, lower = more similar) and a 0–1 similarity score (higher = more similar), enabling threshold-based near-duplicate detection.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "a": {
   "type": "string"
  },
  "b": {
   "type": "string"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "distance": 3,
  "similarity": 0.95
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/simhash-near-duplicate-detection-8e7d3026/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.24klabs.ai](https://www.zero.xyz/host/api.24klabs.ai/llms.txt)