Run DocAI Fabric on your own machine
Self-host the entire platform on a single machine with one Docker Compose file, even a laptop.
You can now run the entire DocAI Fabric platform on a single machine with one Docker Compose file, from a dedicated server down to a laptop for a quick demo. Any operating system that runs Docker works (Linux, Windows, and macOS including Apple Silicon).
Everything runs on the one host: the application, the workflow workers, the built-in OCR engines, the web interface, and a Redis job queue. OCR runs locally with the bundled engines, so no cloud OCR service is needed and no document images leave your machine for text recognition. You connect your own language model for classification and extraction: an on-prem GPU server (Ollama, vLLM, and similar) or a cloud provider, added in the app on the Models page.
All your data lives in one folder, which makes backup and restore a simple file copy, and updates are a pull and restart.
See the standalone deployment guide to get started.
Want updates like this by email?
Occasional highlights, straight to your inbox. Unsubscribe anytime.