DigitalOcean Managed Agents

https://www.digitalocean.com/products/managed-agents

DigitalOcean Managed Agents is a set of services for running AI agents in cloud sessions. Harness Runtime executes agent sessions in dedicated Firecracker microVMs; developers and teams use it with coding harnesses and agent frameworks, while Action Gateway provides managed tool access.

Pricing model Usage-based: per-second actual CPU consumption plus peak memory, storage, egress, checkpoints, templates, and applicable tool charges
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. $0.044 per vCPU-hour for CPU actual use
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. Create: under a couple of seconds generally; resume: about 200 ms; LangGraph cold start: about 20 s
Max runtime The longest a sandbox may run. –
Isolation The isolation technology between sandboxes. Dedicated Firecracker microVM per session
GPUs Whether GPU instances are available for sandboxes. –
Docker Whether Docker containers can run inside the sandbox. –
SDKs TypeScript, Go, Python

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

The product is in Public Preview. Standard OCI container images can be registered as reusable templates, but Docker or Docker-in-Docker support inside the sandbox is not documented.

Sources

Last verified 2026-10-04. Corrections land through the update log.