Block XI · Expert level · Lesson 58
Build your own AI apps: no-code, RAG and self-hosted platforms
Seven GitHub repos for setting up your own ChatGPT, NotebookLM or Lovable with your documents, from AnythingLLM and Dyad with no terminal to Langflow, Dify and RAGFlow with Docker. For each one: installation, license and an example for your business.
- 11 min
- 9 animated slides
- 4-question quiz
- Narrated in 6 languages
What you'll see in this lesson
- Build your own AI apps without depending on anyone
With these repos you set up your own ChatGPT, NotebookLM or Lovable with your documents, on your computer or your server.
- AnythingLLM: your private ChatGPT, no terminal needed
An all-in-one app for chatting with your documents and using agents; more than 67,000 stars because it installs like any other program.
- Dyad: your free Lovable, on your own computer
You describe an app and Dyad generates it and previews it locally; almost 22,000 stars as a free alternative to Lovable, v0 or Bolt.
- Langflow: agents with drag and drop blocks
A visual editor for building AI agents and flows and publishing them as an API or MCP server; more than 155,000 stars.
- Open Notebook: your private NotebookLM with podcasts
A free alternative to Google NotebookLM: notebooks with your sources, chat, notes and podcasts; more than 40,000 stars.
- Open Design: design with agents, like in Claude Design
From a brief to web prototypes, presentations and images that respect your design system; more than 100,000 stars in months.
- Dify: build your own ChatGPT with drag and drop
A platform for building chatbots, agents and flows with your documents and dozens of model providers; more than 158,000 stars.
- RAGFlow: answers with the exact citation from your documents
A RAG engine specialized in complex documents, like tables or scanned PDFs, that answers by citing the source; more than 92,000 stars.
- Summary: which repo to choose for each case
Start with what installs without a terminal and move up to Docker when you need several users, an API or complex documents.
Tools and concepts
- AnythingLLM
- Dyad
- Langflow
- Open Notebook
- Open Design
- Dify
- RAGFlow
Sign in with your Google account to access this lesson and the whole course.
More lessons in this block
- Autonomous agents: 7 GitHub repos for delegating tasks to AI
- Coding agents and skills: the repos everyone installs
- Open-source AI for video, image and audio: 7 GitHub repos
- Scraping and research agents: 7 repos to read the internet with AI
- Train your own model
- Clusters: link several machines for bigger models
- Mega-lists and GitHub repos to learn AI (plus marketing skills)