MicroVM sandboxes for AI agents and LLM-generated code that boot in milliseconds, suspend and resume with state, and run desktops, plus durable orchestration and hosted Git.
Use it now
Keep it with up to 9 other tools you're setting up, then share them all as one link.
Try this first
Ask your coding agent:
- Run Code Sandboxes
Use Tensorlake to create isolated MicroVM sandboxes: ‹what you need›
- Snapshot and Resume
Use Tensorlake to suspend and resume sandboxes: ‹what you need›
- Drive a Sandbox Desktop
Use Tensorlake to launch ubuntu-vnc sandboxes: ‹what you need›
Install the CLI
curl -fsSL https://tensorlake.ai/install | sh
Setup guide: docs.tensorlake.ai/sandboxes/quickstart
Call the API
Auth An API key from Tensorlake
Setup guide: docs.tensorlake.ai/sandboxes/quickstart
Compare
Press Compare to add it to the tray at the bottom of the window, then pick up to 2 others anywhere on the site.
- Access
- APICLI
- Auth
- API Key
- Skills
- 5
Skills
Run Code Sandboxes
Create isolated MicroVM sandboxes, run commands and processes, stream output and copy files in and out for agent code.
Snapshot and Resume
Suspend and resume sandboxes with memory and filesystem preserved, or snapshot and clone them, including running processes.
Drive a Sandbox Desktop
Launch ubuntu-vnc sandboxes and control their desktop from Python or JavaScript for computer-use agents.
Orchestrate Agent Workflows
Deploy serverless, crash-resumable agent applications that coordinate sandboxes for long-running workflows.
Host Agent Git Repos
Create managed Git repositories for agent-generated code, mount them into sandboxes and merge branches via SDK or HTTP.