Open-source LLM app development platform for workflows, RAG pipelines, and agent apps
Other
- TypeScript
- Python
- JavaScript

About Dify
Dify is an open-source platform for building LLM applications from prototype to production. A single visual interface combines AI workflows, RAG pipelines, agent capabilities, and model management, so teams can design, test, and ship AI apps without stitching together separate tools.
Build and test workflows on a visual canvas, then connect hundreds of proprietary and open-source models from dozens of inference providers. A prompt IDE helps craft and compare prompts, RAG handles document ingestion and retrieval, and agents built on Function Calling or ReAct can call from 50+ built-in tools. Logs and performance monitoring let you improve apps using production data.
Every app ships with a matching API, making it straightforward to embed Dify into your own backend. Run the self-hosted Community Edition with Docker Compose, deploy to Kubernetes, or use the hosted Dify Cloud service.
Key features
- Visual canvas for AI workflows and testing
- RAG pipeline with document ingestion and retrieval
- Agent support with Function Calling and ReAct
- 50+ built-in tools for AI agents
- APIs for backend integration
Details
- First released
- 2023
- Self-hosting
- Docker Compose
- Platforms
- Web · Docker
- Deployment
- Cloud · self-hostable
- Model support
- Proprietary and open-source LLMs
- Observability
- Application logs and performance
