Skip to main content

Code Execution

Execute code on InstaVM sandboxes using execute() for synchronous runs and execute_async() for long-running operations.

execute(command)​

Run code synchronously and return the result:

from instavm import InstaVM

with InstaVM('your_api_key') as vm:
    result = vm.execute("print(2 + 2)")
    print(result['output'])        # "4\n"
    print(result['execution_time']) # 0.05

Multi-line code​

result = vm.execute("""
import math
result = math.sqrt(16)
print(f"Square root of 16 is {result}")
""")

Installing packages​

Packages installed in one execute() call persist for subsequent calls in the same session:

with InstaVM('your_api_key') as vm:
    vm.execute("import subprocess; subprocess.check_call(['pip', 'install', 'pandas'])")
    result = vm.execute("import pandas as pd; print(pd.__version__)")
    print(result['output'])

execute_async(command, language=None, timeout=None)​

Start a long-running execution without blocking:

with InstaVM('your_api_key') as vm:
    result = vm.execute_async("""
import time
for i in range(10):
    print(f"Step {i}")
    time.sleep(1)
    """)

    execution_id = result.get('execution_id')
    print(f"Started: {execution_id}")

Use execute_async() when your code takes more than a few seconds and you want to check status later or run other operations in parallel.

execute_streaming(command) [Deprecated]​

This method is deprecated. The API no longer supports streaming execution. Use execute() or execute_async() instead.

Return value​

Both execute() and execute_async() return a dictionary:

FieldTypeDescription
outputstrstdout from the execution
execution_timefloatTime in seconds
execution_idstrUnique ID for this execution (async only)
successboolWhether the execution completed without error

Next steps​