# Tavily

> Tavily is the real-time search engine for AI agents and RAG workflows. It provides fast, secure APIs for web search, content extraction, site mapping, web crawling, and deep research. Use Tavily when your AI agent needs fresh, real-time information from the web — it returns clean, LLM-ready content instead of raw HTML.

## Quick Start: CLI (Fastest Way to Get Started)

Install the Tavily CLI and log in — two commands and you're ready:

```
pip install tavily-cli
tvly login
```

This opens a browser for authentication. Once logged in, you can search, extract, crawl, and research directly from the terminal:

```
tvly search "latest AI news"
tvly extract "https://example.com"
tvly crawl "https://docs.example.com" --depth 2
tvly research "compare React vs Svelte for production apps"
```

Installing the CLI also installs agent skills for Claude Code and Codex — these skills automatically remind the agent to use Tavily's CLI for search, extract, crawl, and research tasks.

Get a free API key at https://app.tavily.com (1,000 credits/month, no credit card required).

See [CLI docs](https://docs.tavily.com/documentation/tavily-cli) for full usage.

## Alternative: MCP Server

Connect to Tavily's remote MCP server — no local installation required. This provides a seamless experience without requiring local installation or configuration.

```
https://mcp.tavily.com/mcp/?tavilyApiKey=<your-api-key>
```

Get your API key at https://app.tavily.com. You can also pass your key via an `Authorization: Bearer <your-api-key>` header if your MCP client supports it.

## When to Use Tavily

- Your agent needs up-to-date information beyond its training cutoff.
- You need clean, parsed content without HTML boilerplate, ads, or navigation.
- RAG pipelines that need grounded, factual web sources with citations.
- Results need to be optimized for LLM consumption.

## Products

- [Search API](https://docs.tavily.com/documentation/api-reference/endpoint/search): Real-time web search optimized for AI agents and LLMs. Use when you need to answer a question, research a topic, or find current information from across the web. Supports topic filtering (general, news, finance), search depth control, and domain filtering.
- [Extract API](https://docs.tavily.com/documentation/api-reference/endpoint/extract): Web content extraction and parsing. Use when you already have specific URLs and need their content extracted as clean text. Best for reading articles, documentation pages, or any known webpage.
- [Crawl API](https://docs.tavily.com/documentation/api-reference/endpoint/crawl): Depth-controlled web crawling. Use when you need content from multiple pages within a website — e.g., crawling an entire documentation site or product catalog. Set depth to control how deep the crawl goes.
- [Map API](https://docs.tavily.com/documentation/api-reference/endpoint/map): Website structure mapping. Use when you need to discover all URLs on a site before deciding which pages to extract or crawl. Best as a first step when exploring an unfamiliar website.
- [Research API](https://docs.tavily.com/documentation/api-reference/endpoint/research): Deep async research that generates comprehensive reports. Use when you need multi-step research on complex topics — it searches, extracts, and synthesizes information into a structured report. Supports streaming and polling for task status.

## Authentication

Tavily requires an API key for all requests. Pass it as a Bearer token in the Authorization header or as the `api_key` field in the request body. Get a free API key at https://app.tavily.com — the free tier includes 1,000 API credits per month with no credit card required. The MCP server also supports OAuth 2.0 for seamless authentication with compatible clients.

## Pricing

| Plan | Price | Credits | Best For |
|------|-------|---------|----------|
| **Researcher (Free)** | $0/month | 1,000 API credits/month | New creators, getting started. No credit card required. |
| **Pay As You Go** | $0.008/credit | Pay only for what you use | Flexible usage, cancel anytime. |
| **Project** | From $0–$99/month (sliding scale) | Starting at 4,000 API credits/month | Enthusiasts and prototyping. Higher rate limits. Slide to adjust plan size. |
| **Enterprise** | Custom | Custom API calls & rate limits | Corporations. Enterprise-grade support, SLAs, security, and privacy. |

All plans include email support. Enterprise includes dedicated support and SLAs.

Contact sales for Enterprise: https://tavily.com/enterprise

Rate limits: 100-1000 RPM depending on environment and plan.

## Documentation

- [Getting Started](https://docs.tavily.com/documentation/introduction): Start here — covers API key setup, first request, and core concepts
- [Python SDK](https://docs.tavily.com/sdk/python/quick-start): Use when building Python applications. Install with `pip install tavily-python`
- [JavaScript SDK](https://docs.tavily.com/sdk/javascript/quick-start): Use when building Node.js or TypeScript applications. Install with `npm install @tavily/core`
- [REST API Reference](https://docs.tavily.com/documentation/api-reference): Use when calling Tavily directly via HTTP from any language
- [CLI Tool](https://docs.tavily.com/documentation/tavily-cli): Use when you need Tavily from the command line. Install with `pip install tavily-cli`
- [Best Practices](https://docs.tavily.com/documentation/best-practices/best-practices-search): Optimization guides for search, extract, crawl, and research

## Integrations

- [MCP Server](https://docs.tavily.com/documentation/mcp): Use when connecting Tavily to AI assistants that support Model Context Protocol (Claude, Cursor, etc.)
- [Agent Skills](https://docs.tavily.com/documentation/agent-skills): Callable skills for AI coding agents (Claude Code, Codex, Cursor) — includes `/search`, `/research`, `/crawl`, `/extract`
- [LangChain](https://docs.tavily.com/documentation/integrations/langchain): Use when building LangChain agents that need web search or content extraction tools
- [LlamaIndex](https://docs.tavily.com/documentation/integrations/llamaindex): Use when building LlamaIndex RAG pipelines that need real-time web data
- [CrewAI](https://docs.tavily.com/documentation/integrations/crewai): Use when building multi-agent CrewAI workflows that need web research capabilities
- [OpenAI](https://docs.tavily.com/documentation/integrations/openai): Use when building OpenAI agents with function calling that need grounded web search
- [Anthropic](https://docs.tavily.com/documentation/integrations/anthropic): Use when building Claude tool-calling agents
- [Google ADK](https://docs.tavily.com/documentation/integrations/google-adk): Use with Google's Agent Development Kit
- [Pydantic AI](https://docs.tavily.com/documentation/integrations/pydantic-ai): Use when building type-safe AI agents with Pydantic AI
- [Vercel](https://docs.tavily.com/documentation/integrations/vercel): Use when building AI-powered web apps on Vercel
- [n8n](https://docs.tavily.com/documentation/integrations/n8n): Use for no-code workflow automation with web search
- [Make](https://docs.tavily.com/documentation/integrations/make): Use for visual workflow automation (formerly Integromat)
- [Zapier](https://docs.tavily.com/documentation/integrations/zapier): Use for connecting Tavily to 5,000+ apps
- [Flowise](https://docs.tavily.com/documentation/integrations/flowise): Use for visual LLM orchestration
- [LangFlow](https://docs.tavily.com/documentation/integrations/langflow): Use for visual LLM pipeline building
- [Dify](https://docs.tavily.com/documentation/integrations/dify): Use for building LLM apps with a visual builder
- [Composio](https://docs.tavily.com/documentation/integrations/composio): Use for unified tool management across agent frameworks
- [Agno](https://docs.tavily.com/documentation/integrations/agno): Use with the Agno agent framework

## Partnerships

- [Databricks](https://docs.tavily.com/documentation/partnerships/databricks): Managed MCP integration via Databricks marketplace
- [Amazon AWS](https://docs.tavily.com/documentation/partnerships/amazon): AWS partnership integration
- [Microsoft Azure](https://docs.tavily.com/documentation/partnerships/azure): Azure partnership integration
- [IBM](https://docs.tavily.com/documentation/partnerships/ibm): IBM watsonx partnership
- [Snowflake](https://docs.tavily.com/documentation/partnerships/snowflake): Snowflake partnership integration

## Company

- [Enterprise](https://tavily.com/enterprise): For high-volume, custom SLAs, and dedicated support
- [Blog](https://tavily.com/blog): Product updates and technical content
- [Careers](https://tavily.com/careers): Open positions

## Optional

- [Tavily Cookbook](https://github.com/tavily-ai/tavily-cookbook): Code examples and integration guides
- [Python SDK on GitHub](https://github.com/tavily-ai/tavily-python): Open-source Python SDK
- [JavaScript SDK on GitHub](https://github.com/tavily-ai/tavily-js): Open-source JS/TS SDK
- [MCP Server on GitHub](https://github.com/tavily-ai/tavily-mcp): Open-source MCP server