Skip to content
Vareon

Vareon Cloud, your cloud or your edge.
One Duplex, wherever the work is.

Run Duplex where your data, your keys and your work need it: managed for you in Vareon Cloud, in a cloud account you own, or on your own devices at the edge, on CPU, beside your edge AI. It is the same Duplex in each, and what it learns moves with you.

Choose by what the work needs.

Step 1 of 4: The quickest start

Duplex at your edge.

Duplex runs on CPU, so it can run on the devices and machines where the work happens, beside the AI already there. It brings runtime intelligence to the device itself: learning, deciding and acting where the work is, with or without a connection.

  • Runs on CPU

    Duplex needs no GPU, so it fits on the processors already in your devices, machines and on-site servers.

  • Beside your edge AI

    Your edge AI takes what is new. Duplex answers what it has learned, so the model is called less.

  • Works offline

    It learns, decides and acts with no connection, and its licence works offline too.

  • Learns where the work is

    Each confirmed result is learned on the device, with no training run, and nothing has to leave.

Start in one. Move when you need to.

What Duplex has learned, and its history, go with it to the next placement.

Step 1 of 5: Start

Start in Vareon Cloud, and Duplex learns from every result your team confirms there.

The same in every placement.

Four promises hold in Vareon Cloud, your cloud and your edge alike.

Step 1 of 4: Same product

Plugins, MCP + hooks and the console behave the same in every placement.

Choose the boundary that fits the work.

Start a pilot in Vareon Cloud. For your cloud or your edge, contact us at hello@vareon.com and we plan the deployment with you.