An open-source (source-available) platform for building and deploying LLM apps with drag-and-drop workflows, RAG and agents
Information updated 2026-10-20, source: official site and public reports. Prices and benefits may change. Check the official site before you pay.
Some information has not yet been verified against official sources. Please check the official site before you pay.
Pros
- Visual workflow builder
- Knowledge base (RAG) pipeline
- MCP support
- Deploy apps as web widgets, APIs or MCP tools
- Free self-hosted Community Edition
- Korean interface
Cons
- The free cloud plan (Sandbox) is small, with 200 message credits and 5 apps
- Paid pricing is per workspace, so it can be expensive for individuals
These are opinions compiled by Pickx from public sources.
Cautions
The Community Edition license is a modified Apache-2.0 license (source-available), so check the conditions for commercial resale or multi-tenant use. The official page shows only annual billing prices (monthly billing prices are based on outside sources)
At a glance
Dify is a platform for building and deploying AI apps by weaving prompts, knowledge bases (RAG), agents and tool calls into visual workflows. It suits practitioners and development teams who want to build in-house chatbots/work automation apps without much coding. You can use it in the cloud or install it yourself with Docker (Community Edition).
Main features/supported AI
| Feature/model | Description | Notes |
|---|---|---|
| Workflow Studio | Visually design, test and deploy the execution path of an LLM app by connecting nodes | Chatflow/workflow forms |
| Knowledge Pipeline (RAG) | Extract/chunk/index files/websites/online documents and test search quality | Storage limits by plan |
| Agents | Create agents through conversation or manual setup, and use them as an app/web widget/workflow node | |
| MCP/marketplace | Install model providers, tools, data sources and MCP integrations as plugins. Apps you build can be published as MCP tools | |
| Supported models | OpenAI, Anthropic, Gemini, xAI, Tongyi and others | Use your own API key after credits run out |
Plans
| Plan | Price (monthly billing / yearly billing) | What you can do/limits | Features |
|---|---|---|---|
| Sandbox | Free | 200 message credits, 5 apps, 1 member, 50 knowledge documents/50MB | For trying out |
| Professional |
| 5,000 message credits per month, 50 apps, 3 members, 5GB knowledge/500 documents | Billed per workspace |
| Team |
| 10,000 message credits per month, 200 apps, 50 members, 20GB knowledge/1,000 documents | About 17% savings with yearly billing |
| Community Edition (self-hosted) | Free | Install yourself with Docker, 1 workspace | Source-available (modified license based on Apache-2.0) |
| Enterprise | Quote | Multiple workspaces, SSO, SLA, security controls, official support | Self-hosted |
Who it suits
- Sandbox: People who want to build an AI workflow/RAG chatbot for the first time
- Professional: When a small team wants to run an in-house knowledge chatbot or work app right away in the cloud
- Community Edition: Developers who must keep data on in-house servers or want to minimize cost
Compared with similar services
- Dify vs n8n: Dify is strong at building LLM apps/RAG chatbots, while n8n is strong at work automation between apps (triggers/integrations)
- Dify vs Make: Make is no-code automation centered on 3,000+ app integrations, while Dify focuses on building AI apps themselves
- Dify vs code frameworks such as LangChain: Dify centers on a visual builder, so non-developers can approach it too
Notes
- Information checked: 2026-10-20 (only the yearly billing price is listed officially, the monthly billing price is still from an external source)
- Dify official pricing
- Dify official homepage
- Monthly billing price reference (external source, vdf.ai)
Good for
8 views (3 today)
Prices converted at 1 USD = ₩1,329.49 (2026-10-19). Original price shown in brackets. Taxes as listed by each provider.
Comments
Sign in to leave a comment.
No comments yet. Be the first.
Your comments and nickname are public. If you see abuse or advertising, report it here and we will review and remove it.