Loading page
Loading page
AI · Microsoft Copilot Studio alternative
Build AI chatbots, agents and RAG workflows on a visual canvas, then publish them as web apps or APIs.
Dify is a platform for building apps on top of large language models. You design a workflow on a canvas: take input, retrieve from your documents, call a model, use a tool, and return an answer. Then you publish it as a chat web app or call it from your own code through an API.
It covers the full loop. There is a RAG pipeline that reads PDFs and slides, a prompt IDE to compare models, agents that run in a sandbox and use tools, and logs so you can see what users asked and fix weak answers.
Dify fits product teams who want to ship AI features without writing all the plumbing. It also fits companies that want internal assistants over their own documents, and developers who want a backend for AI apps that non-developers can also edit.
The self-hosted Community Edition keeps your prompts, documents and chat logs in your own service. You choose which model providers see your data, or use a local model.
The trade-off is size. This template runs many containers, including PostgreSQL with pgvector, Redis, a code sandbox and a plugin daemon, so it needs more memory than most apps. You also pay your model providers directly, and you look after updates and database backups.
Containers: nginx:1.27-alpine, busybox:1.36, langgenius/dify-api:1.15.0, langgenius/dify-web:1.15.0, pgvector/pgvector:pg16, redis:6-alpine, langgenius/dify-sandbox:0.2.15, ubuntu/squid:latest, langgenius/dify-plugin-daemon:0.6.3-local
Step 1
Choose the Starter plan. It costs $20 a month and runs up to 3 apps.
Step 2
Open App templates in the dashboard, choose Dify, and click Deploy. We set up the containers, secrets and domain.
Step 3
Wait a few minutes for the first deploy, then open your new URL and finish the Dify setup screen.