open-source agentic AI data assistant for the next generation of AI + Data products.
About DB-GPT
DB-GPT is an open-source framework for building AI-native data applications. Its main idea is to replace hand-written SQL workflows with natural language, so a person can ask questions of a database and get answers, charts and reports.
The framework includes AWEL, a workflow expression language for chaining steps, a multi-agent layer, retrieval over private data, and a model management layer that works with many open-source and hosted language models. A companion project, DB-GPT-Hub, covers fine-tuning models for text-to-SQL. Recent versions added support for the MCP protocol.
Because it is designed to run in a company's own environment, it is aimed at teams that need to keep data private. It comes from the Eosphoros AI community and is mostly documented in Chinese and English.
Key features
- Natural-language access to databases
- AWEL workflow language for chaining steps
- Multi-agent framework
- Retrieval over private data
- Management of many open and hosted models
- Text-to-SQL fine-tuning through DB-GPT-Hub
- MCP protocol support
- Runs in your own environment
Links
All content and agents from DB-GPT
Frequently asked questions about DB-GPT
- What is DB-GPT?
- open-source agentic AI data assistant for the next generation of AI + Data products. It is listed on botshub under workplace and productivity.
- Is DB-GPT free or open source?
- DB-GPT is tagged as open source on botshub. Check the official site for current terms.
- What can DB-GPT do?
- Key features include natural-language access to databases, aWEL workflow language for chaining steps, multi-agent framework, retrieval over private data and management of many open and hosted models.
- What are alternatives to DB-GPT?
- Other workplace and productivity on botshub include Glean, Sana, UiPath and conversed.ai. See all alternatives to DB-GPT.
