Dify

Dify (LangGenius)

The open-source LLM app platform: build RAG assistants and agent workflows visually, back them with your choice of 100+ models, and run them on your own infrastructure or Dify Cloud.

Product preview

Overview

Dify is what teams reach for when Chatbase is too simple and raw LangChain is too much code: an open-source platform where you visually compose RAG applications and agent workflows, wire in any of 100+ models, and ship them with a generated API and hosted UI. The whole thing runs on your own Docker host — data residency becomes a config decision, not a vendor promise.

It has quietly become the default backend for internal AI tools across startups and enterprises: document-grounded assistants, workflow automations, evaluation and logging included. The visual builder covers 80% of needs; the remaining 20% drops into Python or API calls. For teams that need governed, self-hosted LLM apps without building platform infrastructure from scratch, Dify is the shortest road — and its API-first design means anything you build stays yours.

Pros / Cons

Pros

Self-hosted: data and spend stay yours
Visual builder to production API in hours
Model-agnostic across 100+ providers
RAG, agents, evals and logging in one box

Cons

Ops burden of self-hosting (upgrades, scaling)
Very complex agent graphs hit the builder's ceiling
Cloud pricing is per-seat and adds up on teams

Who it's for

Internal knowledge assistants over company docs

Customer-facing support bots on your own infra

Agent workflows with human approval steps

Compliance-sensitive LLM deployments

Bottom line

The open-source backbone of serious self-hosted LLM apps — start here before you build platform plumbing.

Key Features

Visual workflow & agent builder

RAG pipeline with document ingestion

100+ model providers

Self-hosted or cloud

Backend-as-API for every app

Compare Dify with alternatives

Dify vs Hugging Face Dify vs Replicate Dify vs Ollama 4 products · AI Platforms

Frequently asked questions

What is Dify?

Dify is a AI Platforms tool developed by Dify (LangGenius). The company was founded in 2023 and is headquartered in Remote (open source). The open-source LLM app platform: build RAG assistants and agent workflows visually, back them with your choice of 100+ models, and run them on your own infrastructure or Dify Cloud.