Sandboxes | Render

Introducing Render Sandboxes: a computer for every agent.
Give every agent its own isolated compute, persistent state, and filesystem.
Manage your sandbox’s lifecycle with TypeScript or Python SDKs or CLI.
Each Render sandbox runs in its own Firecracker microVM, so agents can use Docker and return to the processes they left running.
Give your agents tools, files, and processes that persist between sessions.
Run your whole stack with Docker Compose so your agents can test changes end to end.
Keep user-submitted code, model output, and uploaded files isolated from your app’s data and secrets.
+1,204 started · −1,183 stopped in the last minute
Start new sandboxes from custom images or saved snapshots, with the files and tools each task needs.
Stream stdout, stderr, and exit codes so your agent can inspect a failure and try a fix.
Start sandboxes in under a second, even when traffic arrives in bursts.
Render Sandboxes suspend automatically when idle, with no compute charges until your agent sends its next command. While a sandbox runs, you’re billed only for the CPU it actually uses.
Create Render API keys with only the permissions a task needs, like managing sandboxes without access to other services.
Limit outbound requests to approved domains, or block outbound traffic entirely.
Authenticate approved API requests without giving sandbox code the raw keys or tokens.
Copy this prompt into your coding agent to start a sandbox, run a command, and see the output.
Follow the quickstart at https://render.com/docs/sandboxes to run my first Render sandbox. Show me the output, and terminate the sandbox when you’re done.
Your sandbox ran the command and shut down.
Give every agent its own isolated compute, persistent state, and filesystem.