Regex + gazetteer entity extraction with offsets API

Deterministic NER over free text: extract emails, URLs, IPv4 addresses, phone numbers, money amounts, dates, @handles, #hashtags and capitalized organisation/person candidates (Title-Case spans with an org-suffix gazetteer), each with character offsets and a type. A regex+gazetteer pipeline, no LLM. Answers 'extract entities from this text', 'find the emails and urls', 'pull out names and organizations', 'get money and dates with offsets'.

Price$0.01per request

MethodPOST

Route/v1/text/entities

StatusLive

MIME typeapplication/json

Rate limit60/minute

CacheNo cache

textentitiesnernamed-entity-recognitionextractiongazetteernlpregex

API URLhttps://x402.hexl.dev/v1/text/entities

Integration docs

Example request

{
  "text": "Acme Corp hired @jane; email bob@acme.com about $5 million."
}

Example response

{
  "entityCount": 4,
  "countsByType": {
    "email": 1,
    "money": 1,
    "handle": 1,
    "organization": 1
  },
  "entities": [
    {
      "type": "organization",
      "text": "Acme Corp",
      "start": 0,
      "end": 9
    },
    {
      "type": "handle",
      "text": "@jane",
      "start": 16,
      "end": 21
    },
    {
      "type": "email",
      "text": "bob@acme.com",
      "start": 29,
      "end": 41
    },
    {
      "type": "money",
      "text": "$5 million",
      "start": 48,
      "end": 58
    }
  ]
}

Input schema

{
  "type": "object",
  "required": [
    "text"
  ],
  "properties": {
    "text": {
      "type": "string",
      "examples": [
        "Acme Corp hired @jane; email bob@acme.com about $5 million."
      ]
    }
  }
}

Output schema

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