![image of a customer service representative working in a digital interface environment [digital project]](https://cdn.prod.website-files.com/69df88af0a7e746f9df0551f/69df8949b63a10f032693c73_15a8ed48-b275-4740-a0f2-3b7994bd53b9.avif)
Your models. Every device.
Run local AI models seamlessly across your phone, tablet, and desktop — no cloud required.
Run local AI models seamlessly across your phone, tablet, and desktop — no cloud required.
Your models run on your hardware. Your conversations, your context, your data — none of it touches a third-party server unless you choose otherwise.
Designed with a native, fluid experience across iOS, Android, macOS, Windows, and Linux.
Connect your devices into a personal compute mesh. Route inference to whichever node has capacity. Your desktop can serve your phone. Your work machine can serve your home.
core capabilities
Run and manage AI models privately on any device. No cloud. No compromise.
One codebase powering a native experience everywhere. The same model, the same conversation, the same context — across all your screens.
Route model requests across your personal node network. When your phone needs a heavy model, let your desktop handle it — seamlessly.
End-to-end encrypted node communication. No telemetry. No model usage data sent anywhere. What runs locally, stays local.
Pricing & plans
Select a plan to run decentralized AI models on all your devices. Transparent pricing. Cancel anytime.
supported models
Shelley supports the leading open-weight models out of the box. New releases sync automatically. No API keys. No subscriptions per model.
8B · 70B
7B · Small
4B · 12B · 27B
Mini · Full
7B · 14B
7B · 32B
6B · 34B
New models added continuously
Shelley — AI is in active development. Join the waitlist and help shape what local AI on every device looks like.
