Your data, your platform
Run Nyuro in our cloud, or on your own servers with your data kept locally and backed up to the cloud. Either way, you decide when a request may use a frontier model.
Nyuro Cloud
We host everything, so there is nothing to install. The AI Model Router runs in our cloud today; the AI App Builder and AI Employees join it as they leave early access.
Start free →On-prem with cloud backup
Nyuro runs on your own servers. Your documents, conversations and open-source models stay with you, and a backup is kept in the cloud so a hardware failure never costs you your data.
Join the waitlist →What stays where
| Nyuro Cloud | On-prem (early access) | |
|---|---|---|
| Your documents and knowledge | Our cloud | Your servers, backed up to the cloud |
| Conversations and history | Our cloud | Your servers, backed up to the cloud |
| Open-source models | GPUs we run | Your own GPUs |
| Frontier models (OpenAI, Anthropic, Google) | The vendor's cloud, through the AI Model Router | The vendor's cloud, only for the tasks you allow |
Frontier models, only when you choose
Open-source models handle everyday work at lower cost. When a task needs a frontier model, the AI Model Router sends it to the vendor you picked, at their list price with no markup on tokens. You can limit which models each key may use.