Dify is an open-source platform for developing agentic workflows, supporting various AI models and integrations. It offers features like a visual workflow builder, RAG pipeline, and agent capabilities, with deployment options for both cloud and self-hosted environments.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
Dify is an open-source LLM app development platform designed for agentic workflow development. It integrates AI workflows, RAG pipelines, and agent capabilities, enabling users to transition from prototype to production.
How it works
Dify provides a visual canvas for building and testing AI workflows. It supports integration with numerous proprietary and open-source LLMs, including GPT, Mistral, and Llama3. Users can leverage a prompt IDE for crafting prompts and comparing model performance. Dify also features extensive RAG capabilities for document ingestion and retrieval, and allows for the definition of agents with built-in tools like Google Search and DALL·E.
Getting started
To start the Dify server using Docker Compose, follow these commands:
cd dify
cd docker
cp .env.example .env
docker compose up -d
After running these commands, access the Dify dashboard at http://localhost/install to begin the initialization process.
Recent releases
- 1.16.0-rc1 (2026-07-09): Introducing Dify Agent (Experimental): A New Agent Experience in Dify.
- 1.15.0 (2026-06-25): Dify 1.15.0 Release Note with UX & UI enhancements.
- 1.14.2 (2026-05-19): Security fixes, agent groundwork, workflow reliability, and deployment updates.
- 1.14.1 (2026-05-12): Security hardening and workflow stability improvements.
- 1.14.0 (2026-04-29): Collaboration features added.
Traction
Dify has 148523 stars and 23414 forks on GitHub.
Caveats
- Open issues: 953
- License: None listed
- Created on: 2023-04-12