This Python wrapper provides seamless integration with the Tavily API, enabling developers to easily incorporate advanced web search, content extraction, and site crawling capabilities into their applications. It is designed for developers building AI-powered applications, particularly those leveraging Retrieval-Augmented Generation (RAG) or requiring concise, fact-based answers.

How It Works

The library acts as a client for Tavily's REST API, abstracting away HTTP requests and response parsing. It offers distinct methods for different functionalities: search for general web queries, get_search_context for generating RAG-ready text snippets, qna_search for direct question answering, and extract for scraping content from specified URLs. The crawl method, currently in invite-only beta, allows for site traversal based on a starting URL and a search query.

Quick Start & Requirements

• Install via pip: pip install tavily-python
• Requires a Tavily API key.
• Documentation: Python API Reference

Highlighted Details

• Supports generating context for RAG applications with get_search_context.
• Offers a qna_search method for direct question answering, suitable for LLM integration.
• The extract method can process multiple URLs simultaneously and optionally include image data.
• crawl functionality is available in invite-only beta for site traversal.

Maintenance & Community

• Free tier offers 1,000 credits per month for personal use.
• Contact for support: support@tavily.com
• Updates and news via Developer Community.

Licensing & Compatibility

• Licensed under the MIT license.
• Permissive for commercial use and integration with closed-source applications.

Limitations & Caveats

The crawl functionality is currently in an invite-only beta phase, limiting its immediate availability.