Repo LLM Apps Limited info

langgenius/dify

Open-source LLM app platform combining visual AI workflows, RAG pipelines, agents, model management and observability, with cloud, self-hosted and enterprise options.

  • 158k GitHub stars
  • TypeScript
  • ⚖️ NOASSERTION
  • 🎯 Intermediate
langgenius/dify preview image

What it is

Dify is an open-source LLM app development platform. Its interface combines AI workflows, RAG pipelines, agent capabilities, model management and observability features (including Opik, Langfuse and Arize Phoenix). It is meant to take you from prototype to production, and every offering comes with corresponding APIs.

Who it's for

  • Developers building LLM-powered apps who want a visual workflow canvas
  • Teams that need RAG pipelines over documents such as PDFs and PPTs
  • Organizations that want to self-host, or need SSO, RBAC and enterprise support SLAs
  • Developers who want to call LLM app functionality through APIs from their own business logic

Requirements

Requirements

  • CPU >= 2 Core
  • RAM >= 4 GiB
  • Docker installed
  • Docker Compose v2.24.0 or later

Setup

  1. Start Dify with Docker Compose

    From the cloned repository, go to the docker directory, copy the example env file, and start the services. The README gives these commands with cd dify as the first step.

    bash
    cd dify
    cd docker
    cp .env.example .env
    docker compose up -d
  2. Open the dashboard and initialize

    After the containers are running, open http://localhost/install in your browser and start the initialization process.

Pros & cons

Pros

  • Pro:Visual canvas for building and testing AI workflows
  • Pro:Supports hundreds of proprietary and open-source LLMs from dozens of providers, including any OpenAI API-compatible models
  • Pro:Built-in RAG pipeline covers document ingestion to retrieval, with text extraction from PDFs, PPTs and other common formats
  • Pro:All offerings come with corresponding APIs (Backend-as-a-Service), plus LLMOps monitoring and log analysis

Cons

  • Con:Licensed under the Dify Open Source License, which is based on Apache 2.0 with additional conditions rather than plain Apache 2.0
  • Con:Self-hosting requires minimum 2 CPU cores, 4 GiB RAM, and Docker Compose v2.24.0 or later
  • Con:SSO, RBAC security and enterprise support SLAs are available only through Dify Enterprise

Images