tavily-ai/tavily-python

The Tavily Python SDK allows for easy interaction with the Tavily API, offering the full range of our search, extract, crawl, map, and research functionalities directly from your Python programs. Easily integrate smart search, content extraction, and research capabilities into your applications, harnessing Tavily's powerful features.

What it solves

Tavily Python SDK provides a streamlined way for developers to integrate real-time web search, content extraction, and automated research capabilities into their Python applications. It specifically targets the needs of AI agents and RAG (Retrieval-Augmented Generation) systems that require accurate, fact-based context from the live web to reduce hallucinations and improve answer quality.

How it works

The SDK acts as a wrapper for the Tavily API, allowing users to perform several types of operations:

  • Search: Executes web queries to find information, including a specialized get_search_context method for RAG applications and a qna_search for direct answers.
  • Extract: Pulls raw content and images from a list of specific URLs (up to 20 at once).
  • Crawl: Traverses a website starting from a base URL to find pages matching specific instructions.
  • Map: Visualizes and discovers the structure of a website based on a base URL and instructions.
  • Research: Generates comprehensive research reports with automatic source gathering and analysis, supporting both standard and streaming output.

Who it’s for

Developers building AI agents, LLM-powered applications, and RAG pipelines who need a reliable way to fetch high-quality, cleaned web data for their models.

Highlights

  • RAG-Optimized: Dedicated methods for generating context strings and concise Q&A answers specifically for LLMs.
  • Flexible Authentication: Supports a keyless mode for quick testing, API keys for full access, and custom HTTP sessions for enterprise proxy gateways.
  • Comprehensive Web Intelligence: Combines search, extraction, crawling, and site mapping into a single interface.
  • Enterprise Ready: Includes session and user tracking identifiers (project_id, session_id, human_id) for request attribution.

Related

  • Dispatch
    Tavily Deep Research: Achieving State of the Art in AI Research Agents

    Tavily has developed a state-of-the-art research agent that reduces token consumption by 66% compared to Open Deep Research by implementing a linear context engineering approach based on human-like information distillation.

  • Project
    oxylabs/oxylabs-ai-studio-py

    A Python SDK for Oxylabs AI Studio, providing AI-powered tools for web crawling, scraping, and agentic browsing to extract structured data using natural language prompts.

  • Project
    DeepXiv/deepxiv_sdk

    DeepXiv SDK is a Python/CLI toolkit that lets you ask natural‑language questions and receive answers grounded in real arXiv papers or cached web pages, with genuine citations. It offers progressive reading, fine‑grained filters, multiple effort levels, and a simple `Reader` class for integration into autonomous agents. Free auto‑registered tokens give general access; registered accounts unlock 30 agentic calls per day. MIT‑licensed.

  • Project
    infiniflow/ragflow

    An open-source RAG engine that combines deep document understanding and agentic capabilities to create high-fidelity, production-ready AI systems from complex unstructured data.

  • Project
    deepgram/deepgram-python-sdk

    The official Python SDK for Deepgram, providing easy access to automated speech recognition, text-to-speech, and language understanding APIs.