Managed isolated environments for Python

0.25.7 · active · verified Tue Apr 14

Isolate offers a pluggable end-to-end solution for building, managing, and using isolated environments (virtualenv, conda, remote, and more). While directly installable, its primary utility for most users is realized as an underlying component of the fal.ai platform, enabling serverless execution of Python functions with specified dependencies in isolated cloud environments. The current version is 0.25.7, and the project appears to have an active release cadence, with version 0.25.6 uploaded to PyPI in early April 2026.

Warnings

gotcha The `isolate` library (fal-ai/isolate) is primarily an internal component of the `fal.ai` platform. Most users will interact with its functionality indirectly through the `fal` Python library and CLI, rather than directly importing and using `isolate` classes or functions. Direct API stability for standalone use may not be a high priority for its maintainers.
Fix: For isolated execution of Python functions, especially in a serverless context, refer to the `fal` (fal-ai/fal) library's documentation and quickstarts. Install `fal` and use its provided decorators and client for managing environments.
gotcha Confusion with other similarly named libraries (e.g., `isolated-environment`, `run-isolated`) is possible due to the generic nature of the name. Ensure you are referencing `fal-ai/isolate` when investigating this specific project.
Fix: Always verify the PyPI project page (pypi.org/project/isolate) and the official GitHub repository (github.com/fal-ai/isolate) to confirm you are using the correct library and its documentation.
gotcha While `isolate` provides core isolation capabilities, deploying functions to remote or cloud environments (as often intended with the `fal` ecosystem) requires authentication. Users will need to set up `FAL_KEY` environment variables or use `fal auth login` to authenticate their `fal` client.
Fix: Ensure that the `fal` CLI is installed and configured with appropriate authentication credentials (`fal auth login` or `export FAL_KEY=...`) when deploying or running functions remotely via `fal`.

Install

pip install isolate Install PyPI package

Imports

Isolate functionality is typically accessed via the 'fal' library
```
import fal
```
The 'isolate' library is primarily an internal dependency of the 'fal' platform (fal-ai/fal). Direct, standalone API usage examples for 'isolate' are not commonly documented for end-users; its core functionality for creating and managing isolated environments is usually exposed through the 'fal' client and its decorators (e.g., @fal.function).

Quickstart

This quickstart demonstrates how the `fal` library, which utilizes `isolate` internally, allows users to define and run Python functions in isolated environments with specified dependencies. The `@fal.function` decorator handles the environment setup, including dependency installation, making the underlying isolation seamless to the developer. Direct usage of the `isolate` library's API is not a common pattern for end-users.

import fal
import os

@fal.function(
    "virtualenv",
    requirements=["pyjokes"],
    # For authentication, fal_client typically picks up FAL_KEY from environment variables
    # or 'fal auth login'. For this example, we assume it's set.
    # For a runnable quickstart, replace os.environ.get with your actual key if testing locally
    # fal_client is not directly used here, but implicitly by fal.function for deployment.
)
def tell_joke() -> str:
    import pyjokes
    joke = pyjokes.get_joke()
    return joke

# To run this, you would typically use the fal CLI after saving to a file (e.g., my_app.py):
# fal run my_app.py::tell_joke
# The output would be the joke returned by the function, executed in an isolated environment.

# This code snippet demonstrates how 'fal' (which uses 'isolate' internally)
# enables running a Python function with specific dependencies in an isolated environment.
# Direct invocation of 'isolate' classes/functions is not a common user pattern.
print('Function defined. To run, use `fal run <filename>::tell_joke` after `pip install fal` and `fal auth login`.')

view raw JSON →