Free🇭🇰
Open-source RAG from the University of Hong Kong that builds a knowledge graph from documents and answers questions via graph and vectors.
DB-GPT overview and key facts (October 2026)
Open-source AI data assistant with agents that generates SQL, runs Python analysis and builds reports and dashboards.
Free
What is DB-GPT?
DB-GPT is an open-source agentic data assistant that connects to databases and files, answers questions by writing SQL and Python, and turns the results into reports and dashboards, with the option to run on local models.
DB-GPT connects to databases, CSV and Excel files and knowledge bases. It generates SQL autonomously, executes code in a sandbox and runs reusable skills for domain tasks. It supports models such as DeepSeek, Qwen, GLM, Llama and Gemma, locally through vLLM or llama.cpp or through OpenAI-compatible APIs. Installation works via PyPI, Docker or source. The README is available in English, Chinese and Japanese. The GitHub API showed about 20,000 stars and a push on September 28, 2026.
DB-GPT AI features and use cases
- Queries databases in plain languageUsers ask a question and DB-GPT writes and runs the SQL. Business teams get answers from operational databases without waiting for an analyst.
- Analyzes spreadsheets with codeFor CSV and Excel files the assistant writes Python and executes it in a sandbox. Controllers or analysts can explore files without writing code themselves.
- Produces reports and dashboardsResults are turned into reports and dashboards, and skills package recurring analyses. Regular evaluations can be repeated with one request.
What sets DB-GPT apart
- Combines text to SQL with sandboxed Python execution and reusable skills in one assistant.
- Strong support for Chinese open models such as Qwen, GLM and DeepSeek alongside Llama and Gemma.
Where DB-GPT is strongest
Available internationally
Open-source software with no country restriction that runs on your own infrastructure anywhere.
Questions to ask before you buy DB-GPT
Give DB-GPT read-only database accounts at first, since generated SQL and code run against your data. Decide whether data may go to a cloud model or must stay on local models. The project has no commercial support contract, so plan internal ownership for updates.
DB-GPT reviews
No reviews yet. Be the first to share how your team uses it.
Every review shows its verification level. We read each one before it goes live. How we check reviews
Questions about DB-GPT
Is DB-GPT free?
Yes, it is released under the MIT license.
Can DB-GPT run without a cloud model?
Yes. The README describes local deployment of models via vLLM or llama.cpp.
Which data sources does it support?
Databases, CSV and Excel files and knowledge bases, according to the README.
DB-GPT update historyno updates yet
No changes recorded for DB-GPT yet.
Changes the vendor made and confirmed user reports, each checked against the vendor's own pages. We review every tool monthly.
Alternatives to DB-GPT
Tools in the same category (Data & analytics) with the most overlapping markets.
Free🇨🇳
Open-source document parser from OpenDataLab at Shanghai AI Laboratory: turns PDFs, scans and Office files into Markdown or JSON for AI.
Free🇺🇸
Open-source Python library that builds scraping pipelines with LLMs: describe the data you want and it extracts it from pages or files.
FreeAvailable internationally
Open-source Python crawler that turns websites into Markdown for RAG and AI agents, with an optional hosted API and MCP access.