What makes OpenJet a Jetson LLM?

OpenJet is built for Jetson-class hardware and local inference workflows. It is designed to run on NVIDIA Jetson devices while keeping logs, model execution, and approvals on the device.

Which Jetson devices does OpenJet target?

OpenJet is aimed at Jetson Nano, Xavier NX, Orin Nano, Orin NX, and AGX Orin environments.

Why use a Jetson LLM instead of a hosted model?

A Jetson LLM keeps execution close to the hardware, reduces cloud dependency, and makes more sense when logs, tools, and operators already live on the device.

jetson llm

Jetson LLM for NVIDIA Jetson devices

If you are specifically looking for a Jetson LLM, this page is about running local models on NVIDIA Jetson hardware rather than sending operational context to a hosted API.

OpenJet gives you a terminal-native Jetson LLM workflow for Nano, Xavier NX, Orin Nano, Orin NX, and AGX Orin. It keeps inference, logs, and approvals on the device, while exposing a practical interface for local operations on constrained Jetson hardware.

Install OpenJet Read the docs

Why local wins

Built for Jetson-class hardware. OpenJet is designed around Jetson deployment realities: limited RAM, local storage, operator access through terminal workflows, and device-side execution.
Works across the Jetson line. The workflow is aimed at Jetson Nano, Xavier NX, Orin Nano, Orin NX, and AGX Orin deployments where on-device inference matters more than remote convenience.
Local model plus local actions. OpenJet keeps the model, logs, and approved actions on the Jetson, which is a better fit for field systems and embedded deployments than cloud-first agents.

Common deployment paths

Deploy a Jetson LLM on Orin devices for local diagnostics and ops workflows.
Run local models on embedded NVIDIA hardware without routing logs through a cloud provider.
Use a terminal-native LLM interface on Jetson systems that need human-approved execution.

Frequently asked questions

What makes OpenJet a Jetson LLM?: OpenJet is built for Jetson-class hardware and local inference workflows. It is designed to run on NVIDIA Jetson devices while keeping logs, model execution, and approvals on the device.
Which Jetson devices does OpenJet target?: OpenJet is aimed at Jetson Nano, Xavier NX, Orin Nano, Orin NX, and AGX Orin environments.
Why use a Jetson LLM instead of a hosted model?: A Jetson LLM keeps execution close to the hardware, reduces cloud dependency, and makes more sense when logs, tools, and operators already live on the device.

Explore OpenJet

Embed a local model with the Python SDK, run Qwen3.8-27B as a local coding agent, compare local coding-agent alternatives, read the docs, or go back to the homepage.