sandbox.watch

Agent37 Cloud

https://www.agent37.com/cloud

Agent37 Cloud provisions a persistent, isolated instance for an AI agent, accessible through an API and its own URL. Developers use it to run an agent for each end user, using a supplied agent template or a custom image.

Pricing model Cloud API: prepaid, per-minute billing for sized vCPU and RAM while awake, plus disk until deletion; browser-managed hosting has separate monthly subscriptions.
Pricing The headline compute price as the provider states it. Sorting ranks prices stated per vCPU over time, converted to hourly. Other pricing models sort after, unranked. Cloud API: $0.0011 per vCPU-hour
Free while idle Whether a stopped, paused, or sleeping sandbox costs nothing. ✕
Elastic Resources flex with what the sandbox actually uses, and the bill follows: yes when the provider bills on active or observed use, no when capacity is reserved or allocated and billed while it runs. ✕
Memory snapshots Whether RAM state survives a pause and resume, not just disk. ✓
Wake on request Whether a stopped sandbox wakes automatically on inbound traffic. ✓
Start / resume Typical time from create or resume to running. Sorting ranks stated times, converted to seconds; claims with no figure sort after, unranked. A few seconds to restore from sleep; create-to-ready time is not stated.
Max runtime The longest a sandbox may run. No fixed limit; an instance lives until you delete it
Isolation The isolation technology between sandboxes. gVisor
GPUs Whether GPU instances are available for sandboxes. –
Docker Whether Docker containers can run inside the sandbox. ✕
SDKs –

✓ yes · ✕ no · – no cited public fact.

A resumed instance can require a fresh boot if its checkpoint cannot be restored; a running status does not necessarily mean the agent is ready.

Sources

Last verified 2026-10-05. Corrections land through the update log.