Individual Lookup

Run one web-search-backed research prompt and get the answer back directly. Synchronous.

POST /research_with_ai

Costs 1 credit, charged only when the model returns text. Returns the answer directly in the response, no list and no polling.

Pass a complete, self-contained prompt. Unlike the List Lookup route, there is no row to substitute into, so {{column_name}} placeholders are not expanded here, send the finished question text.

Headers

Name	Value
Content-Type	application/json

Body

Name	Type	Description
api_key	string	Your API Key.
prompt	string	The fully-resolved prompt. No {{column}} substitution happens on this endpoint.

Example

Input

{
  "api_key": "your_api_key",
  "prompt": "What is Anthropic's most recent funding round as of 2026, and how much did they raise?"
}

Output

{
  "status": "success",
  "data": {
    "research_with_ai": "Anthropic's most recent funding round as of 2026 was its Series H round, announced on May 28, 2026. It raised $65 billion at a $965 billion post-money valuation."
  },
  "credits_used": 1,
  "credits_remaining": 4823,
  "success": true
}

No answer

{
  "status": "not_found",
  "data": {
    "research_with_ai": null
  },
  "credits_used": 0,
  "credits_remaining": 4823,
  "success": true
}

Error

{
  "status": "error",
  "error": {
    "code": "validation_error",
    "message": "prompt: Field required"
  },
  "success": false
}

Response

Field	Type	Description
research_with_ai	string	The model's answer. Same name as the list column this endpoint writes to.

The record is under data. See Response Format for the fields every endpoint returns.

To run a prompt template against every row of a list in one call instead, use List Lookup.