DeepBI is an AI-native data analysis platform designed for users seeking to derive insights and make data-driven decisions through conversational interaction. It empowers individuals to explore, query, visualize, and share data from various sources using natural language, effectively acting as an LLM-powered data scientist.

How It Works

DeepBI leverages large language models (LLMs) to interpret user queries in natural language. It translates these conversational requests into executable data queries and generates visualizations. The platform supports assembling these visualizations into interactive dashboards, with future plans for automated data analysis reports. This LLM-centric approach aims to democratize data analysis, making complex data exploration accessible without requiring deep technical expertise.

Quick Start & Requirements

• Windows: Download window_install_exe_EN.zip from tags and run the .exe.
• Docker: Clone the repo (git clone https://github.com/DeepInsight-AI/DeepBI.git), cd DeepBI, and run ./Install.sh. Access via http://ip:8338.
• Ubuntu: Requires Redis, PostgreSQL (v16 recommended), and Python 3.8.x. Clone the repo and run . ubuntu_install.sh.
• Supported Databases: MySQL, PostgreSQL, Doris, StarRocks, CSV/Excel Import, MongoDB.
• Platforms: Windows (WSL 22.04+ recommended), Linux (Ubuntu 20.04/22.04 tested), macOS (12.7+ tested).
• Resources: Minimum 1 core/2GB RAM, 2 cores/4GB RAM recommended.

Highlighted Details

• Conversational data analysis and query generation via dialogue.
• Supports multiple data sources including popular SQL databases and file imports.
• Multi-platform support for Windows, Linux, and macOS.
• Internationalization with support for Chinese and English.

Maintenance & Community

The project encourages community support via stars and forks. Contact information for developers (dev@deepbi.com) and business inquiries (hi@deepbi.com) is provided.

Licensing & Compatibility

The repository does not explicitly state a license in the README. Users should verify licensing terms before commercial use or integration into closed-source projects.

Limitations & Caveats

Automated data analysis reports are listed as "to be developed." Windows 10 users may require version 22H2 or higher for WSL compatibility. The README does not specify the LLM models used or their hosting requirements.