Practical overview
When it fits
Use E2B when an AI-agent workload needs an isolated machine for code execution or tool use, and choose between E2B cloud hosting and hosting in a customer’s cloud account.
Who it is for
- Teams building AI agents that execute code or use tools.
- Developers needing isolated compute for agent sessions or background jobs.
Documented uses
- Provide an isolated environment for an agent to run code.
- Provision a per-session sandbox for agents that use computers and tools.
- Supply temporary compute for background agent jobs.
- Provide sandbox infrastructure for reinforcement-learning rollouts.
Requirements & limits
The pricing page lists a free Hobby tier and paid upgrades. E2B describes itself as an open-source project and offers an API.
- The service provisions a microVM for each session.
- Hosting is available through E2B’s cloud or a customer’s cloud account.
- The listed free Hobby tier and paid upgrades may matter when evaluating access.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Create a sandbox and run a command with the Python SDK
Start an E2B sandbox and run a shell command from Python.
Before you start
- Install the E2B Python SDK with
pip install e2b. - Set the
E2B_API_KEYenvironment variable.
Steps
Install the SDK with the command below.
pip install e2bSet
E2B_API_KEYin the environment.Create a sandbox with
Sandbox.create().Run a command with
sandbox.commands.run('echo "Hello from E2B Sandbox!"')and print the result’sstdout.
Expected outcome
A sandbox is created and the command output is printed.
Guide limitations
- The documented example demonstrates a single command; it does not cover sandbox cleanup or error handling.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 6 October 2026
- Original website title
- E2B | The Enterprise AI Agent Cloud · Official homepage title · retained in its source language