dify
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move
What is dify?
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.
Overview of dify
What it is
Dify is an open-source LLM app development platform. It provides an intuitive interface that combines AI workflow, RAG pipeline, agent capabilities, model management, and observability features (including Opik, Langfuse, and Arize Phoenix). The platform allows users to go from prototype to production quickly. It can be deployed on cloud, VPC, or self-hosted.
Key features
- Workflow: Build and test powerful AI workflows on a visual canvas.
- Comprehensive model support: Seamless integration with hundreds of proprietary/open-source LLMs from dozens of inference providers and self-hosted solutions, covering GPT, Mistral, Llama3, and any OpenAI API-compatible models.
- Prompt IDE: Intuitive interface for crafting prompts, comparing model performance, and adding features like text-to-speech to chat-based apps.
- RAG Pipeline: Extensive RAG capabilities from document ingestion to retrieval, with out-of-box support for text extraction from PDFs, PPTs, and other common document formats.
- Agent: Autonomous agents with a sandbox to run commands, install software, and handle files. Agents can be given skills and connected to tools from Dify Marketplace, MCP servers, or custom APIs.
- LLMOps: Monitor and analyze application logs and performance over time, with continuous improvement of prompts, datasets, and models based on production data.
- Backend-as-a-Service: All offerings come with corresponding APIs for easy integration into business logic.
Use cases
Building agentic workflows and RAG pipelines. Developing AI applications that require multi-model support, prompt engineering, and observability. Teams can move from prototype to production without rebuilding the stack. Suitable for both cloud and self-hosted environments.
Getting started
Before installing, ensure minimum system requirements: CPU >= 2 Core, RAM >= 4 GiB. The easiest way is via Docker Compose (v2.24.0 or later):bashcd difycd dockercp .env.example .envdocker compose up -dAfter running, access the dashboard at http://localhost/install and start initialization.
License and pricing
This repository is licensed under the Dify Open Source License, based on Apache 2.0 with additional conditions. The platform is open source. Cloud and Enterprise editions are available; pricing details are not specified in the provided materials.
Based on the project's public README and repository metadata.