Dify Open Source

DomesticFree + PaidAgent Workflow

Visual workflow, knowledge base Q&A, and agent building

Developer:Suzhou Yuling Artificial Intelligence Technology Co., Ltd.

Dify Open SourceIntroduction

Dify is a leading open-source visual platform for building AI agents and applications. It features core capabilities including Workflow orchestration, RAG (Retrieval-Augmented Generation), and prompt management, with support for private deployment to ensure full data ownership—making it a top choice for technical teams. Balancing low-code visual canvas design with advanced customization, Dify lets you drag-and-drop nodes to construct multi-step agent workflows or write custom code nodes and integrate proprietary APIs. It supports a wide range of popular LLMs beyond any single vendor, allowing flexible switching between models like Doubao, Qwen, GLM, GPT, and more. Ideal for building enterprise internal chatbots, document processing agents, and business process automation assistants. The open-source version can be self-deployed on your own servers; the Enterprise edition offers official SaaS hosting and commercial support. With an active community and comprehensive documentation, Dify is one of the most popular open-source AI application engines globally. Note: This summary is for reference only; please verify critical details manually.

Key Features of the Dify Open Source Platform

Key Product Features

1Visual Agent Workflow Orchestration
2RAG Knowledge Base: Document Parsing and Retrieval-QA
3Supports integration with multiple mainstream large language models
4Code Node, Custom API Call
5Complete On-Premises Deployment Solution
6Prompt Version Management, A/B Testing
7Application Access API, Conversation Log Audit

Dify Open Source Pricing

Free Tier, Paid Plans, Billing Options, Private Deployment & Enterprise Pricing

Billing Type:Free + Paid

1Free quota: The SaaS version includes a free tier; the open-source version can be self-deployed.
2Paid: Commercial Enterprise Edition and Technical Support Services
3Billing: Subscription
4Open-source version is free; compute costs are borne by the user.
5Private Deployment: Fully supports on-premises deployment

Dify Open Source Usage Guide

Entry URL, Registration Requirements, Brief Operation Steps, and API / Client Support

Entry Point:dify.ai

1Entry: https://dify.ai
2Register: Self-hosted deployment requires no platform account
3SaaS version: Register with phone number or email
4Steps: Get the open-source code and complete deployment
5Create an app and select workflow or conversation type
6Orchestrate nodes, integrate models, and connect tools
7Upload Knowledge Base
8Debug, deploy, and integrate with the business system

Dify Open SourceParameter Overview

Information compiled from public sources; official information prevails.

Is data used for training?: Private deployment keeps data under your control; SaaS allows disabling training.
Underlying model: Compatible with virtually all major large language models
Context length: Follow the access model limit
Multi-modal support: Text, Images, Documents
Search Online: Supports search plugin integration
File Parsing: PDF, Word, Excel, TXT, Markdown
Enterprise Edition and Deployment: No public information available
Compliance Certification: No public information available
Domestic Availability: No public information available