---
title: "Tavily Search and Extract API"
description: "Tavily Search and Extract API on looot as of 2026-09-28: 4 jobs through 4 endpoints, how you reach it, what it costs, and how to run it."
---

{/* Generated by scripts/generate-providers.mjs from data/providers.json. Do not edit. */}

<ProviderHeader id="tavily" />

## What it's good for

Tavily Search and Extract API is in Web data, SEO/AEO. Reach for it to search the web and get ranked results; get extract; get crawl.

As of 2026-09-28 (catalog revision 85280) it serves 4 jobs through 4 endpoints.

## Jobs

| Job id | What it does | Endpoints |
| --- | --- | --- |
| [`crawl.get`](/catalog/web-data) | Get crawl | 1 |
| [`extract.get`](/catalog/web-data) | Get extract | 1 |
| `map.get` | Get map | 1 |
| [`web.search`](/catalog/seo-aeo) | Search the web and get ranked results | 1 |

## Access

- **Runs on looot's key** for 4 endpoints: nothing to set up, the run spends your looot balance. You can connect your own key instead; see [Bring your own key](/guides/bring-your-own-key).

## Price basis

1 per call (from $0.008), 3 per result (from $0.0008). A call that fails at the provider costs $0; see [What's free](/money/free-and-failures).

## Free first in fallback

Partly. 3 endpoints bill per result, so a miss costs nothing and fallback can move them ahead of flat-priced rows. See [Free first](/concepts/free-first).

## Run it

A job run takes shared input names, not Tavily Search and Extract API's own field names; looot maps them for each provider. For `job:crawl.get`, `url` is taken by 1 of its 1 providers. See [Job inputs](/concepts/job-inputs).

Example input for `looot run` or the MCP [`run`](/mcp-tools/run) tool, with placeholder values:

<CodeGroup>

```bash CLI
looot run job:crawl.get --input '{"url":"https://example.com"}'
```

```json MCP
{
  "endpointId": "job:crawl.get",
  "input": {
    "url": "https://example.com"
  },
  "idempotencyKey": "<new unique key>"
}
```

</CodeGroup>

A job run picks among all the job's providers. To run Tavily Search and Extract API itself, name one of its endpoints (MCP [`search`](/mcp-tools/search) with `{"filters": {"provider": "tavily"}}` lists them), for example `tavily-crawl`: `looot inspect tavily-crawl` prints its exact input fields and price.

## In the live catalog

- Browse: [looot.ai/public-catalog/tavily-crawl](https://looot.ai/public-catalog/tavily-crawl), or search [looot.ai/public-catalog](https://looot.ai/public-catalog?text=Tavily%20Search%20and%20Extract%20API) for Tavily Search and Extract API.
- Every endpoint, no account needed: `curl "https://api.looot.ai/v1/public-catalog?provider=tavily"`

Tavily Search and Extract API's name and logo are trademarks of their owner, shown to identify the provider.

<Related />