AI Search Data API Quickstart

Before you begin

You need an API key. Create one on the API keys page of the dashboard. New accounts start with 500 free credits. API plans and credits covers pricing.

1. Set your API key

export QUERYING_API_KEY="vt_..."

Keep the key on your server. See Authentication.

2. Submit a task

Ask Gemini a question. taskType picks the engine; payload.prompt is what you ask.

curl -X POST https://api.querying.ai/v1/async/task \
  -H "Authorization: Bearer $QUERYING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "taskType": "GEMINI", "payload": { "prompt": "best wireless earbuds for running", "country": "US" } }'

The API accepts the task and queues it. For example:

{
  "success": true,
  "task": {
    "id": "8f2c1e40-...",
    "taskType": "GEMINI",
    "status": "QUEUED",
    "priority": 1,
    "createdAt": "2026-10-19T04:12:00.000Z"
  },
  "credits": { "creditsToCharge": 0, "creditsCharged": 0 }
}

credits is zero on submit. The task’s price is held from your balance, captured when it completes, and released if it fails. See Credits and limits.

Chat engines such as GEMINI read payload.prompt. GOOGLE, AIMODE, the Naver engines and GOOGLE_SERP read payload.query. Each falls back to the other field, so either one works. See Which payload field is read.

3. Collect the result

Webhook (recommended)
Polling

Add webhook.url when you submit. When the task finishes, querying.ai POSTs the result to that URL, so you never poll.

curl -X POST https://api.querying.ai/v1/async/task \
  -H "Authorization: Bearer $QUERYING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "taskType": "GEMINI",
    "payload": { "prompt": "best wireless earbuds for running", "country": "US" },
    "webhook": { "url": "https://your-server.example.com/hook" }
  }'

Your endpoint receives { task, credits, response }. Every delivery is signed with a key derived from your API key; verify the signature before you trust the body. See Webhooks for signing, retries and the payload.

Omit webhook and poll the task by id. response is absent until the task is COMPLETED or FAILED.

curl https://api.querying.ai/v1/async/task/8f2c1e40-... \
  -H "Authorization: Bearer $QUERYING_API_KEY"

For example, a completed Gemini task:

{
  "success": true,
  "task": { "id": "8f2c1e40-...", "taskType": "GEMINI", "status": "COMPLETED", "...": "..." },
  "credits": { "creditsToCharge": 1.25, "creditsCharged": 1.25 },
  "response": {
    "text": "For running, look for earbuds with a secure fit and sweat resistance ...",
    "sources": [{ "position": 1, "url": "https://example.com/running-earbuds", "label": "..." }]
  }
}

On a terminal task, credits shows what the task settled at: the engine’s price when it completed, 0 when it failed. While the task is QUEUED or PROCESSING, both values are 0.

A finished task can be polled for 24 hours. After that it returns 404 RESOURCE_NOT_FOUND. See Async tasks.

Next steps

Pick an engine

Every taskType, what it returns, and what it costs.

Batches and grids

Submit up to 500 tasks per request, or 10,000 as one batch.

Webhooks

Payloads, signature verification and retries.

Credits and limits

Prices, concurrency and daily caps.

Get started

Core concepts

Answer engines

Search and web data

Request options

Research

GEO monitoring

Account and data

AI Search Data API Quickstart

Before you begin

1. Set your API key

2. Submit a task

3. Collect the result

Next steps

Pick an engine

Batches and grids

Webhooks

Credits and limits

​Before you begin

​1. Set your API key

​2. Submit a task

​3. Collect the result

​Next steps

Pick an engine

Batches and grids

Webhooks

Credits and limits

Before you begin

1. Set your API key

2. Submit a task

3. Collect the result

Next steps