---
title: "Can I run AI in an air-gapped or on-premises environment?"
description: "Yes. Open-weight models and local inference software run with no internet connection. Download the models and software on a connected machine, move them into the secure network, and run everything there."
url: https://exolabs.net/faq/air-gapped-on-premises-ai/
updated: 2026-10-26
---

# Can I run AI in an air-gapped or on-premises environment?

Yes. Open-weight models and local inference software run with no internet connection. Download the models and software on a connected machine, move them into the secure network, and run everything there.

Verify the files you bring in. Download weights from the official publisher, record checksums, and prefer the safetensors format, which cannot run code when loaded.

Turn off remote code. Some models ask to run custom Python from their repository (trust_remote_code). Leave this off unless you have reviewed the code.

Plan updates. New models and security patches need the same transfer process, so schedule it.

**Where exo fits:** exo runs offline, can read models from read-only shared storage (EXO_MODELS_READ_ONLY_DIRS), and keeps trust_remote_code off by default. Namespaces keep separate clusters on one network apart.

Sources:
- [exo on GitHub (README)](https://github.com/exo-explore/exo)
- [Hugging Face: safetensors](https://huggingface.co/docs/safetensors/index)

Page: https://exolabs.net/faq/air-gapped-on-premises-ai/ · Updated 2026-10-26 · Deployments: https://exolabs.net/#business (sales@exolabs.net)