Cognee

🚀 Cognee Project Introduction

Cognee is a tool for building and managing AI memory, which helps developers handle natural language processing and knowledge graph-related tasks more efficiently. With Cognee, you can easily transform text information into structured knowledge graphs and interact with multiple large language models (LLMs).

🚀 Quick Start

Quick Navigation

• Project Documentation
• GitHub Repository
• Demo Video
• Join Discord

✨ Features

• Multimodal Support: Handles not only text but also various data types such as images and audio.
• Knowledge Graph Construction: Automatically extracts entities and relationships to build a structured knowledge network.
• Extensible Plugin System: Extends functionality through plugins, supporting various custom processing logics.
• Distributed Storage: Supports large-scale data storage and efficient querying.

📦 Installation

Install Cognee using pip:

pip install cognee

💻 Usage Examples

Basic Usage

Environment Configuration

Set environment variables (it is recommended to use the .env file):

import os
os.environ["LLM_API_KEY"] = "YOUR_OPENAI_API_KEY"

For more configuration methods, please refer to the Documentation.

Example Code

Simple Example

Here is a simple example of using Cognee:

import cognee
import asyncio


async def main():
    # Add text to Cognee
    await cognee.add("自然语言处理 (NLP) 是计算机科学和信息检索的交叉学科领域。")

    # Generate a knowledge graph
    await cognee.cognify()

    # Query the knowledge graph
    results = await cognee.search("告诉我关于 NLP 的知识")

    # Display the results
    for result in results:
        print(result)


if __name__ == '__main__':
    asyncio.run(main())

Example Output

自然语言处理 (NLP) 是一个跨学科领域，涉及计算机科学和信息检索。它关注于机器如何理解和处理人类语言。

Visualization of the knowledge graph: Open the Link in your browser to view the full effect.

For more advanced usage, please refer to the Documentation.

📚 Documentation

Demo Cases

1. What is AI Memory: Learn about Cognee
2. Simple GraphRAG Demo: GraphRAG Example
3. Integration of Cognee and Ollama: Local Model Example

Code of Conduct

Please abide by the Code of Conduct to jointly create a friendly open-source community environment.

Star History

Check the project's Star History to understand the community's activity.

Welcome to join the Cognee open-source community and build a better AI memory tool together!