# Open-Jet: self-hosted Agentic TUI for air-gapped Jetsons

**URL:** <https://forums.developer.nvidia.com/t/open-jet-self-hosted-agentic-tui-for-air-gapped-jetsons/362093>\
**Category:** Jetson Projects\
**Tags:** generative\_ai, agentic-ai, jetson\
**Created:** [March 2, 2026, 11:12am UTC](https://forums.developer.nvidia.com/t/open-jet-self-hosted-agentic-tui-for-air-gapped-jetsons/362093 "2026-03-02T11:12:32Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![flouisdev](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@flouisdev](https://forums.developer.nvidia.com/u/flouisdev)\
**Post date:** [March 2, 2026, 11:12am UTC](https://forums.developer.nvidia.com/t/open-jet-self-hosted-agentic-tui-for-air-gapped-jetsons/362093/1 "2026-03-02T11:12:32Z")

</div>

I am building a Terminal User Interface (like Claude Code) for self-hosted AI agents on Jetsons. Works in air-gapped environments.  
Unlike other solutions, this is optimised for unified memory machines, as to avoid OOM errors.

The agent can do stuff like edit, read, create files - manage and interpret data locally.

Currently, it gets ~17 tok/s on Jetson Orin Nano 8GB using Qwen3-4B-Instruct-4bit  
In the future, adding TensorRT .engine support which will boost inference further. I am trying to get the memory footprint down, so if anyone has knowledge on kv cache optimisation, that would be great.

I would love to get your feedback and people try running it on more capable devices and models - post your results here.

Run

```auto
pip install open-jet
open-jet --setup

```

Website: [https://www.openjet.dev/](https://www.openjet.dev/)  
Directly on PyPi: [open-jet · PyPI](https://pypi.org/project/open-jet/)