Open-source BI tool for drag-and-drop data visualization, dashboards and sharing, positioned as an alternative to Tableau.
- Stars
- 20,178
- Forks
- 2,979
- Open issues
- 457
- Last commit
- 5 hours ago
What is it?
DB-GPT is an open-source agentic AI data assistant aimed at teams that want to ask questions of their data in natural language. It connects to relational databases, CSV and Excel files, warehouses, and knowledge bases, then lets an LLM plan multi-step tasks, write and run SQL and Python, and return charts, dashboards, HTML reports, and written summaries.
Beyond the assistant itself, DB-GPT is a platform for building data agents, workflows, and applications using its agent framework, AWEL workflow orchestration, and RAG support. Reusable skills package domain knowledge and analysis methods so they can be loaded for repeatable business analysis. Code and tools run in isolated sandboxed environments.
The project supports a range of LLM providers, including DeepSeek, Qwen, GLM, and Llama models, and can be installed through a one-line script, PyPI (dbgpt-app), or Docker.
Key features
- 01Converts natural language questions into SQL and runs queries against connected databases.
- 02Analyzes CSV and Excel files with Python and generates charts and visual reports.
- 03Produces database profiling and financial analysis reports.
- 04Loads reusable skills that package domain analysis workflows.
- 05Executes code and tools inside sandboxed environments.
- 06Provides AWEL workflow orchestration and RAG for building custom data applications.
Use cases
- Letting business users ask questions of a sales or operations database without writing SQL.
- Running exploratory analysis on CSV or Excel exports and getting a shareable HTML report.
- Building an internal AI data assistant that combines databases and knowledge base documents.
- Packaging a team's recurring analysis methods as skills that others can reuse.
Quick start
- 1
Install DB-GPT from PyPI with uv (recommended)
uv pip install dbgpt-app - 2
Or install DB-GPT from PyPI with pip
pip install dbgpt-app - 3
Start DB-GPT and follow the interactive setup wizard to choose an LLM provider and enter your API key
dbgpt start - 4
Open the web UI at http://localhost:5670
Things to know
- The one-line installer script is documented only for macOS and Linux.
- Using a hosted model requires an API key from an LLM provider.
Topics
FAQ
- What is DB-GPT?
- An open-source agentic AI data assistant that writes SQL and code over your databases and files, runs skills in sandboxes, and produces reports. DB-GPT is an open-source agentic AI data assistant aimed at teams that want to ask questions of their data in natural language. It connects to relational databases, CSV and Excel files, warehouses, and knowledge bases, then lets an LLM plan multi-step tasks, write and run SQL and Python, and return charts, dashboards, HTML reports, and written summaries.
- Is DB-GPT free to use? What license does it have?
- DB-GPT is open source under the MIT license. Check the license terms before using it in commercial products.
- Is DB-GPT still maintained?
- Actively maintained. The last commit to DB-GPT was on Oct 22, 2026.
- What can I use DB-GPT for?
- Letting business users ask questions of a sales or operations database without writing SQL. Running exploratory analysis on CSV or Excel exports and getting a shareable HTML report. Building an internal AI data assistant that combines databases and knowledge base documents. Packaging a team's recurring analysis methods as skills that others can reuse.
- How do I install DB-GPT?
- Install DB-GPT from PyPI with uv (recommended): uv pip install dbgpt-app Or install DB-GPT from PyPI with pip: pip install dbgpt-app Start DB-GPT and follow the interactive setup wizard to choose an LLM provider and enter your API key: dbgpt start Open the web UI at http://localhost:5670
- Where can I get DB-GPT?
- The source code and installation instructions are on GitHub: https://github.com/eosphoros-ai/DB-GPT
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source desktop app that analyzes your exported chat histories from WhatsApp, LINE, QQ, Discord and more, with local SQL and AI agents.
An open-source general AI agent framework that runs LLM-driven agents from a terminal, with tool use, MCP support and browser automation.
Open-source GenBI engine that lets AI agents turn natural-language questions into governed SQL and shareable dashboards over 20+ data sources.
A minimal AI software engineering agent that solves GitHub issues or helps in the command line, using only bash and about 100 lines of core Python.
A deep research skill and pipeline for Claude Code and OpenAI Codex that gathers web and scholarly sources into a persistent, searchable markdown vault and writes cited reports.