Agent 37 Cloud vs E2B

Agent 37 Cloud

Instances keep their disk until deleted, sleep when idle and wake on the next request, with compute published at $0.80 a vCPU-month and stable error codes on both APIs. The product launched on 29 September 2026. No status page, OpenAPI file, SDK library or published rate limits were found, and the starter credit needs a card.

E2B

Firecracker microVM with its own kernel per sandbox. Two major incidents over an hour in September 2026, on sandbox creation and on creating from snapshots.

Agent 37 Cloud

1. Send the key as Authorization: Bearer to api.agent37.com and as X-Agent37-Key to the instance URL. The Bearer form on instance URLs is deprecated
2. Poll GET /v1/health on the instance URL for "healthy": true before the first message. status: running only means the machine is up
3. Check whether error is a string before reading error.code. Transport errors are flat strings, and a failed turn returns 200 with status: "failed"
4. Don't blindly retry an exec that timed out after 13 minutes. The command keeps running, so a retry starts a second copy
5. Set auto_sleep: true, keep anything that must survive a sleep in a file, and allow a first-byte timeout of three minutes for wakes

E2B

1. Set a timeout when you create a sandbox. The default is 5 minutes
2. Pause rather than kill when you'll come back. Resume takes about a second and nothing is billed while paused
3. Use Secret.fill in network transforms instead of passing API keys into the sandbox environment
4. Pace sandbox creation. Hobby allows 1 a second and 20 running at once
5. Move to the v2 sandbox endpoints. SDK 2.51.0 and later use them by default