Maritime vs Vercel Sandbox

Maritime

Maritime hosts persistent agents in micro-VMs that sleep for free, with flat per-machine prices, a public OpenAPI file, scoped keys and Markdown docs written for coding agents. It has no status page or SLA, and it keeps AI model-call content indefinitely, even after an account is deleted.

Vercel Sandbox

Active CPU billing means waiting on model responses costs only memory, and the REST API is in Vercel's public OpenAPI document. Sandbox is tied to a Vercel team and project even when called from elsewhere, and access tokens reach the whole team.

Maritime

1. Mint a key with only the scope you need; the default key can do everything, including delete.
2. Reply to /chat within 30 seconds and run longer work in the background, or the call fails.
3. Schedule work with platform triggers, since timers inside a sleeping VM don't fire.
4. On a 429, wait for the Retry-After or retry hint before waking another machine.
5. Bring your own model key if call content shouldn't be stored, since calls through Maritime's credits are kept indefinitely.

Vercel Sandbox

1. Call sandbox.stop() when the task is done. Memory bills until the session ends
2. Use Sandbox.getOrCreate with a name so retries land in the same sandbox
3. Set networkPolicy to deny-all for untrusted code. The default is allow-all
4. Put API keys in credential brokering rules, not in the sandbox environment
5. Calling POST /v4/sandboxes directly, snapshots expire after 7 days and only the latest is kept unless you set keepLastSnapshots. The SDK still uses v3, with 30 days