# Dify

Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.

Repository: https://github.com/langgenius/dify
Canonical: https://ross.abutalabs.com/products/dify
Homepage: https://dify.ai
Language: TypeScript
License: NOASSERTION
License Family: other
Topics: ai, gpt, llm, openai, python, agent, nextjs, workflow, genai, automation, low-code, mcp, no-code, agentic-ai, agentic-framework, agentic-workflow, claude, skills, deepseek
Last push: 2026-08-26T21:09:50+00:00

## Health v2 (maintenance only)
Score: 93/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 88
- inputs: {"age_days": 1239, "days_push": 7, "days_rel": 8, "gap_med": 7.0, "n_releases_24m": 73}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 153595, forks 24273 (observed 2026-08-28T04:12:24.390142+00:00)

## What it is
Dify is an open-source LLM application development platform for building agentic workflows and RAG pipelines in a collaborative visual workspace. It supports many AI models and tools and can be deployed on cloud, VPC, or self-hosted infrastructure.

## Use cases
- build ai agent workflows without code
- create rag chatbot over my documents
- self-host an llm app platform
- prototype and deploy llm applications
- build ai customer support bot
- orchestrate multiple llm tools in one pipeline
- compare and switch between gpt and claude models

## When to choose
- you want a visual, low-code workspace to build LLM agents and RAG apps
- your team needs to move from prototype to production without rebuilding the stack
- you need self-hosted or VPC deployment with broad model support
- non-developers and developers must collaborate on AI workflows

## When to avoid
- you need fine-grained programmatic control over agent internals that a low-code platform abstracts away
- you only need a lightweight SDK or library to embed in an existing codebase
- you require a permissive open-source license for commercial redistribution (license is custom)
- your use case is simple single-prompt inference with no workflow needs

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, rag, workflow-automation, chatbot, llm-inference, prompt-engineering, mcp, web-framework
- domain: artificial-intelligence, large-language-models, chatbots, self-hosted
- platform: self-hosted, cloud, python
- tags: low-code, no-code, llm-app-platform, visual-workflow, genai, openai, claude, deepseek, ai-agents, retrieval-augmented-generation, automation, docker, web-server, nodejs

## Member repositories
- langgenius/dify (main) score 93
- langgenius/dify-sandbox (infra) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:24.390142+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:11:00.199038+00:00, confidence not recorded.
  - readme: https://github.com/langgenius/dify (fetched 2026-08-28T04:12:24.390142+00:00, sha 180fab4e3e98)
- Data as of 2026-08-30T08:39:29.467469+00:00.