RAG・知識ベース
SQLBot
SQLBot — DataEase's conversational analytics: ask in natural language, get SQL answers via LLMs plus RAG.
★ 6.9k スターGPL-3.0
ガイドを見る →DB-GPTは、eosphoros-aiが公開しているオープンソースのRAG・知識ベースプロジェクトです（GitHub: eosphoros-ai/DB-GPT）。GitHubスターは20,090、フォークは2,957で、MITライセンスの下で公開されています。GitHubでの最終更新は2026年10月4日です。
An agentic AI data assistant - chat with your databases, run SQL and code, get reports
Decide which database or files the assistant should analyze first.
Install the dbgpt-app package from PyPI with uv or pip - no source checkout required.
Start DB-GPT and let the first-run wizard choose your LLM provider and API key.
Open the web UI and ask a natural-language question about your data.
An agent explores your schema, plans an approach and iterates instead of answering in one shot.
Generates and runs SQL plus Python code to answer questions and build analyses.
Connects to multiple databases and data sources from one assistant.
Add skills and agents to extend what the assistant can do.
Code runs in a sandbox, keeping autonomous execution contained.
Turns analysis into reports through the built-in web interface.
Get the source code from GitHub and enter the project directory.
git clone https://github.com/eosphoros-ai/DB-GPT.git
cd DB-GPTThe recommended install uses uv; plain pip works too. Python 3.10+ is required.
uv pip install dbgpt-app
# or: pip install dbgpt-appFirst run launches an interactive wizard for your LLM provider, then starts the web server.
dbgpt start
# then open http://localhost:5670It is maintained by eosphoros-ai and contributors on GitHub.
Yes - DB-GPT is MIT licensed and free to use and modify.
The setup wizard asks for an LLM provider API key (profiles include OpenAI, Kimi/Moonshot and MiniMax in the install scripts); self-hosted models are an alternative for advanced setups.
Compared with text-to-SQL point tools, DB-GPT is a full agentic assistant - planning, code execution, skills and reporting - closer to an AI data analyst than a SQL generator.
Run uv pip install dbgpt-app, then dbgpt start, complete the wizard, and open the web UI at localhost:5670.
SQLBot — DataEase's conversational analytics: ask in natural language, get SQL answers via LLMs plus RAG.
Self-hosted NotebookLM alternative — chat with documents, get audio summaries.
Complete API layer for private AI apps on local models: RAG, skills, tools, MCP and text-to-SQL
Build Neo4j knowledge graphs from unstructured data using LLMs — then chat with your graph.
Ready-to-run templates for RAG, AI pipelines and enterprise search that stay in sync with live data
DeepSearcher — open-source deep research over your private data, with reasoning models and Milvus retrieval.
LibreAIカタログに掲載されたRAG・知識ベースプロジェクトの中で最もスターの多いもの — DB-GPTが合わなかった場合に検討したい、現実的な候補リストです。
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or…
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context…
Lightweight OCR toolkit turning PDFs and images into structured, LLM-ready data in 100+ languages
MinerU — turns PDFs and Office documents into LLM-ready Markdown and JSON for agents and RAG pipelines, Apache-2.0 plus terms
Stop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience
カタログの中でDB-GPTとスター数が近いRAG・知識ベースプロジェクト — 前後の棚もそのまま見て回れます。
Vector similarity search inside Postgres - store embeddings next to your data
A private AI platform for agents, assistants and enterprise search over your documents
Turn PDFs into AI-ready data: extract Markdown, JSON and tables — and auto-tag for accessibility
A blazing-fast vector index in Rust, built on TurboQuant - with Python bindings
A knowledge-base platform with visual AI workflows — build Q&A systems on your own documents
Air-gapped, open-source NotebookLM alternative that turns local documents into briefings, decks and podcasts.