How to beginner · 3 min read

How to use E2B with OpenAI

Q: How to use E2B with OpenAI

Use the e2b_code_interpreter package to create a Sandbox instance with your OpenAI API key from os.environ. Run code securely by calling sandbox.run_code() and manage files with sandbox.files.write(). Always close the sandbox with sandbox.close() to release resources.

Quick answer

Use the e2b_code_interpreter package to create a Sandbox instance with your OpenAI API key from os.environ. Run code securely by calling sandbox.run_code() and manage files with sandbox.files.write(). Always close the sandbox with sandbox.close() to release resources.

PREREQUISITES

Python 3.8+
OpenAI API key (set in environment variable OPENAI_API_KEY)
pip install e2b-code-interpreter>=1.0
pip install openai>=1.0

Setup

Install the e2b-code-interpreter package and ensure your OpenAI API key is set in the environment variable OPENAI_API_KEY. This package enables secure sandboxed code execution integrated with OpenAI.

bash

pip install e2b-code-interpreter openai

output

Collecting e2b-code-interpreter
Collecting openai
Successfully installed e2b-code-interpreter-1.0.0 openai-1.0.0

Step by step

Create a Sandbox instance using your OpenAI API key from os.environ. Use sandbox.run_code() to execute Python code securely. You can write files into the sandbox environment with sandbox.files.write(). Always call sandbox.close() to clean up resources.

python

import os
from e2b_code_interpreter import Sandbox

# Initialize sandbox with OpenAI API key from environment
sandbox = Sandbox(api_key=os.environ["OPENAI_API_KEY"])

# Run simple Python code
result = sandbox.run_code("print('Hello from sandbox')")
print(result.text)

# Write a file into the sandbox
sandbox.files.write("data.csv", b"id,value\n1,100\n2,200")

# Run code that reads the file
code = """
with open('data.csv') as f:
    print(f.read())
"""
file_result = sandbox.run_code(code)
print(file_result.text)

# Close sandbox to release resources
sandbox.close()

output

Hello from sandbox
id,value
1,100
2,200

Common variations

Use sandbox.run_code() for synchronous execution; async support is not currently provided.
Install additional Python packages inside the sandbox by running sandbox.run_code("import subprocess; subprocess.run(['pip','install','package'])").
Use different OpenAI models by configuring the Sandbox instance if supported by the e2b-code-interpreter version.

Troubleshooting

If you see API key missing errors, ensure OPENAI_API_KEY is set in your environment before running the script.
If code execution hangs or fails, verify network connectivity and that your API key has sufficient quota.
Always call sandbox.close() to avoid resource leaks.

✅

Key Takeaways

Use Sandbox from e2b_code_interpreter with your OpenAI API key for secure code execution.
Write files into the sandbox environment using sandbox.files.write() before running code that depends on them.
Always close the sandbox with sandbox.close() to free resources and avoid leaks.

Verified 2026-04 · gpt-4o, gpt-4o-mini

Verify ↗