Using Private Workers

Run tasks on your own hardware.

In the workspace

Toggle to "Private" when running an app:

1Run on: [○ Cloud] [● Private]

Via API

1result = client.run({2    "app": "my-app",3    "input": {...},4    "infra": "private"5})

Use infra: "private_first" when you want private workers when available, with automatic cloud fallback:

1result = client.run({2    "app": "my-app",3    "input": {...},4    "infra": "private_first"5})

Specific workers

Target exact workers:

1result = client.run({2    "app": "my-app",3    "input": {...},4    "infra": "private",5    "workers": ["my-server-gpu-0"]6})

Agents on private

Agents can use private workers too.

When an agent calls a tool, it respects your infra setting.

Monitoring

Check your engines at Engines:

• Online/offline status
• Resource usage
• Running tasks
• Drain / Update — graceful maintenance (see Updating)

To drain or upgrade an engine without killing in-flight tasks, see Managing engines.

Use Drain before maintenance or Update for a safe binary upgrade (in-flight tasks finish first). The same operations are available via the Engines API or client.engines.drain() / client.engines.updateBinary() in the JavaScript SDK.

Caching

The engine caches:

• App code
• Downloaded models
• Container images

Second runs are much faster.

Next

→ Back to Docs