Skip to main content
Self-hosting AppBlips means running it entirely on your own machine or server — no account, no cloud sync, and no data leaving your environment. It’s designed for a single local user, so there’s no sign-in flow and no project history saved anywhere beyond your machine. All you need is an API key for an OpenAI-compatible provider and a few minutes to get up and running.

​
Prerequisites

Self-hosting AppBlips requires Node.js 20.19 or newer (or Node.js 22.12 or newer). npm is included with Node.js, so no separate install is needed. Download Node.js at nodejs.org.

​
Set up with Node.js

1

Clone the repository

Download the AppBlips source code and move into the project folder:
2

Install dependencies

3

Copy the config file

Create your local configuration file from the provided example:
4

Set your AI provider details

Open the newly created .env file in any text editor and fill in the three required variables:
Replace your-api-key with your actual key. The other values shown are the defaults for OpenAI — adjust them if you are using a different provider or model.
5

Start the server

AppBlips starts the development server and prints the local address to your terminal.
6

Open AppBlips in your browser

Navigate to http://localhost:5175 — AppBlips is ready to use. Type a prompt and start building.

​
Using Docker Instead

If you prefer a more stable, longer-lived setup, Docker is a great alternative to npm run dev. The Docker image builds the static client and serves everything — including the builder API — from a single lightweight Node server. Run the following command in the project root (with your .env file in place):
AppBlips will be available at http://localhost:3000. See the Docker setup guide for details on configuring the image, rebuilding after config changes, and running in production.
In self-hosted mode, the /api/chat endpoint has no authentication. Every request that reaches it is treated as the same local user and can spend your configured AI budget. Do not expose AppBlips beyond localhost without putting your own access control — a reverse proxy with authentication, a VPN, or similar — in front of it first.

​
Compatible AI Providers

AppBlips works with any provider that implements the OpenAI chat completions API. You are not limited to OpenAI itself:
  • OpenAI — use https://api.openai.com/v1 with a standard sk-... key
  • Local models via Ollama — point APPBLIPS_LLM_BASE_URL at your Ollama endpoint (e.g. http://localhost:11434/v1) and set APPBLIPS_LLM_MODEL to any model you have pulled
  • Other compatible providers — any service that speaks the OpenAI chat completions format works; update the base URL, key, and model name to match
The provider credentials you set in .env are used only server-side and are never exposed to the browser.

​
What’s Next

Configuration Reference

Explore every .env variable — rate limits, AI mode for generated apps, tuning options, and more.

Docker Setup

Learn how to build and run the Docker image, manage environment variables, and set up a production-ready deployment.