MinHash signature API

Produces a deterministic MinHash signature of a text's shingles for scalable set-similarity / LSH deduplication. Answers 'What is the MinHash signature of this text?', 'How do I estimate Jaccard at scale?'.

Price$0.02per request

MethodPOST

Route/v1/retrieval/minhash-signature

StatusLive

MIME typeapplication/json

Rate limit120/minute

Cache0s public

minhashlshsignaturededupnear-duplicatejaccardshinglerag

API URLhttps://x402.hexl.dev/v1/retrieval/minhash-signature

Integration docs

Example request

{
  "text": "the quick brown fox jumps",
  "numHashes": 8
}

Example response

{
  "signature": [
    2749182080,
    986512237,
    1285380490,
    164582266,
    1359591442,
    1380426985,
    459412572,
    735421094
  ],
  "numHashes": 8,
  "shingleSize": 2,
  "shingleCount": 4,
  "note": "estimate Jaccard via fraction of matching signature positions"
}

Input schema

{
  "type": "object",
  "required": [
    "text"
  ],
  "properties": {
    "text": {
      "type": "string"
    },
    "numHashes": {
      "type": "integer"
    },
    "shingleSize": {
      "type": "integer"
    }
  }
}

Output schema

{
  "type": "object",
  "additionalProperties": true
}