DB-GPT
DB-GPT is an open-source agentic AI data assistant that connects to databases, files, and knowledge bases, enabling natural language queries, autonomous SQL and code generation, and sandboxed execution. It helps developers and teams build AI-native data analysis workflows, generate reports, and create reusable skills. It is designed for those who need to combine LLMs with structured and unstructured data sources for decision-ready insights.
✨ Key features
- Agentic data analysis with task planning and tool use
- Autonomous SQL and Python code generation and execution
- Multi-source data access: databases, CSV/Excel, documents, knowledge bases
- Reusable skills for domain-specific workflows
- Sandboxed code execution for safety
- Generates charts, dashboards, HTML reports, and summaries
🎯 Use cases
- Analyze CSV/Excel files and generate visual reports
- Connect to databases and produce profiling reports
- Ask business questions in natural language and get SQL automatically
- Perform financial report analysis with code and charts
- Build AI-native data assistants for teams or products
📦 Installation
🧰 Requirements: Python 3.10+ and uv or pip for PyPI install; API key for LLM provider (e.g., OpenAI, Moonshot, MiniMax) required for cloud models.
Quick Start (macOS & Linux)
One-line installer:
curl -fsSL https://raw.githubusercontent.com/eosphoros-ai/DB-GPT/main/scripts/install/install.sh | bash
With profile and API key:
curl -fsSL https://raw.githubusercontent.com/eosphoros-ai/DB-GPT/main/scripts/install/install.sh \
| OPENAI_API_KEY=sk-xxx bash -s -- --profile openai
Install via PyPI
uv pip install dbgpt-apppip install dbgpt-app🚀 Usage
After installation, start the server:
dbgpt start
On first run, an interactive setup wizard will guide you through choosing an LLM provider and entering your API key. Once complete, the web server starts automatically.
Open the Web UI at http://localhost:5670.
❓ FAQ
What are the prerequisites for installing via PyPI?
You need Python 3.10+ and uv (recommended) or pip. The default installation includes core framework, OpenAI-compatible LLM support, DashScope/Tongyi, RAG parsing, and ChromaDB.
How do I start DB-GPT after installation?
Run dbgpt start in your terminal. On first run, an interactive setup wizard will guide you through LLM provider selection and API key entry, then the web server starts automatically.
Can I use local models with DB-GPT?
Yes, the README mentions support for local GPU models (vLLM, llama.cpp) and provides links to advanced installation docs for Docker and source code setup.
What data sources can DB-GPT connect to?
It can connect to relational databases, CSV/Excel files, documents, and knowledge bases, and supports mixed-source workflows.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.