Dify

The platform for production-ready agentic workflows. Build AI apps, RAG pipelines, and agents visually with a workflow studio and deploy via cloud, VPC, or open-source Docker.

About Dify

Dify is an open-source and cloud-hosted platform designed to democratize AI agent development by combining powerful AI/ML capabilities with a visual, no-code interface. It allows teams to construct agentic workflows, orchestrate RAG (Retrieval-Augmented Generation) pipelines, and integrate rich tools in one collaborative workspace.

The platform is designed to scale from rapid prototyping to enterprise production without requiring teams to rebuild their tech stack. Developers and citizen builders can visually define how an application thinks, retrieves data, makes decisions, and asks for human input, maintaining full visibility over the execution path.

Key Features

Workflow Studio: A visual builder to turn complex prompt logic into visible execution paths, making it easy to test, debug, and hand off.
Knowledge Pipeline: Create repeatable RAG pipelines to extract, clean, chunk, and index documents for accurate AI retrieval.
Plugin Marketplace: Install model providers, custom data sources, and MCP (Model Context Protocol) integrations to reuse across applications.
Model API Load Balancing: Bypass API rate limits by configuring multiple API keys for model load balancing.
Publish Anywhere: Ship the same visual logic as a hosted web app, an embed, an API endpoint, or an MCP-compatible tool.
Deployment Flexibility: Available via managed Dify Cloud, self-hosted Dify Enterprise (VPC), or the open-source Community Edition via Docker.

Pricing

Sandbox (Free): $0 Includes 200 message credits, 1 workspace, 5 apps, 50 knowledge documents (50MB storage), and standard execution priority.
Professional: $59/month per workspace ($118 savings annually) Includes 5,000 message credits/month, 3 team members, 50 apps, 500 documents (5GB storage), priority processing, and unlimited log history.
Team: $159/month per workspace ($318 savings annually) Includes 10,000 message credits/month, 50 team members, 200 apps, 1,000 documents (20GB storage), top priority processing, and unlimited workflow triggers.

Use Cases

Scaling item generation with multi-stage review for assessments
Governed expert-level AI assistants for life sciences
Unifying banking teams on one shared AI platform
RAG-based knowledge bases for multi-campus services

Pros & Cons

Pros:

Highly visual interface makes debugging AI logic transparent.
Deep flexibility to run open-source via Docker or managed SaaS.
Built-in Model API load balancing prevents rate-limit bottlenecks.

Cons:

Free sandbox is heavily restricted on vector storage (50MB).
App runtime data analysis (Langsmith/Langfuse) is locked behind higher tiers.

Integrations

Notion, Langsmith, Langfuse, Docker, MCP

FAQ

Can I try Dify without a paid subscription?

Yes, you can try Dify for free under the Sandbox plan to experience core features until resource limits (like vector space or app count) are reached.

How do message credits work?

Message credits allow you to test models from OpenAI, Anthropic, Gemini, xAI, and Tongyi. Different models consume different amounts of credits (e.g., GPT-4 uses 20 credits, Claude Haiku uses 5). Once exhausted, you must supply your own API key.

Is Dify open-source?

Yes, Dify offers a Community Edition that is source-available under an Apache-2.0-derivative license, which can be deployed via a single Docker command.