DB-GPT

DB-GPT is an open-source AI-native data application development framework with 16,000+ GitHub stars, providing infrastructure for building database-connected AI applications. The platform features AWEL (Agentic Workflow Expression Language) for orchestrating complex multi-step database operations, a multi-model management layer supporting local and API-based LLMs, and RAG framework optimized for database contexts. Supports PostgreSQL, MySQL, SQLite, DuckDB, ClickHouse, and more. Key capabilities include automated schema understanding, multi-agent collaboration for complex queries, knowledge graph integration, and fine-tuning pipelines for domain-specific accuracy. The platform provides both a web UI for end users and Python APIs for developers building custom applications. Page should cover: AWEL workflow language tutorial, multi-model configuration, RAG implementation details, agent architecture explanation, deployment options (Docker, Kubernetes), and comparison with Vanna.ai and Wren AI.