Dify
Self-hostable platform for building LLM apps, workflows and RAG
What it is
A web platform for building LLM applications that combines a visual workflow builder, RAG pipelines, agent tooling, model management and observability hooks. It is deployed with Docker Compose and also offered as a hosted service. For teams that want a shared workspace instead of writing an LLM app from scratch.
Good for
- Build a RAG chatbot over company documents
- Prototype an agent workflow visually
- Manage models and prompts in one workspace
Install or run
docker compose up -dCommand as given in the project README. Read the repository before you run anything on your machine.
What you need
Docker with Compose v2.35+, at least 2 CPU cores and 4 GiB RAM; model provider keys
Sources
- https://github.com/langgenius/dify (project README, read 10 Nov 2026)
- https://api.github.com/repos/langgenius/dify (GitHub API, read 13 Nov 2026)
The summary is ours, written from the README. Spotted an error? See how we work.