Otto · Hardware

The agent computer.

Plug in. Tap. Deploy. No terminal, no configuration, no waiting.

Otto One

Everything you need.

Otto One handles credentials, files, and tasks locally — inference runs in the cloud, which means faster models, zero maintenance, and a lower price. As local inference matures, you’ll have the option to flex into it. You still own the device. You still own your data.

Intel N95 · 8GB DDR4 · 256GB SSD. Small, silent, always on.

Otto One

8GB · 256GB

$599

Intro pricing

Otto One

16GB · 512GB

$799

Every otto device

NFC tap to pair

Hold your phone. Done.

otto OS pre-installed

Runtime, Gateway, Passkey — ready out of the box.

No terminal required

Plug in, tap, deploy. That's it.

otto app included

Manage everything from your pocket.

Otto Pro

For local inference.

Same tap-to-pair, same otto OS, same app — but now your agent can run models locally. Otto seamlessly routes between local and cloud inference, letting you burst into the cloud when you need it and stay local when you don’t.

Meet Otto Pro

Otto Pro 32GB

ProcessorAMD Ryzen AI 9 HX 370
Memory32GB DDR5
Storage512GB SSD
Local inference8B–30B quantized models
CPU12 cores · 24 threads

Smaller local models, personal workflows, and cloud support when you need it.

$1,699

Order now

Otto Pro 64GB

ProcessorAMD Ryzen AI 9 HX 370
Memory64GB DDR5
Storage1TB SSD
Local inference30B–70B quantized models
CPU12 cores · 24 threads

Larger local models with room for the rest of your agent’s work.

$2,199

Order now

Otto Pro 96GB

ProcessorAMD Ryzen AI 9 HX 370
Memory96GB DDR5
Storage2TB SSD
Local inference70B+ quantized models
CPU12 cores · 24 threads

Larger models and more demanding local workflows.

$2,699

Order now

Model size, quantization, and context length affect memory use and speed. More RAM adds capacity; it does not automatically make the same model faster.

Local inference, managed by otto

Otto Pro doesn’t just give you local compute — it manages it. Otto OS automatically selects the right model and routing for each task: run locally when latency or privacy matters, burst to cloud when you need a 200B model or parallel throughput. You don’t configure this. It just works.

Ready to get started?

Otto One · $599. Ships with otto OS and the app.

Order now