---
id: simple-parsing
version: "0.1.9"
license: MIT License Copyright (c) 2019 Fabrice Normandin Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to… (full text in the JSON record)
license_treatment: permissive
maintenance: active
---

# simple-parsing — A small utility to simplify and clean up argument parsing scripts.

License: permissive · Maintenance: active · Downloads: 4.3M/mo

## What it is and what it does

simple-parsing wraps Python's argparse to let you define command-line arguments as dataclasses instead of repeated add_argument() calls. It generates help text from docstrings and field comments, handles nested and inherited dataclasses, and supports serialization to json/yaml. The core idea is reducing boilerplate: instead of copy-pasting argument definitions, you define an Options dataclass once and reuse it with automatic prefixing for multiple argument groups.

You use it by creating a dataclass with typed fields, then passing it to ArgumentParser.add_arguments() or calling simple_parsing.parse() directly. It depends on docstring-parser to extract help text from comments and typing-extensions for type annotation support. The package requires Python 3.9 or later.

Use it for:

- Machine learning projects where you need to parse training, validation, and test configurations with identical structure but different prefixes.
- Reusing argument definitions across multiple CLI tools in the same codebase without copy-pasting argparse boilerplate.
- Generating clean, auto-documented --help output from dataclass field comments and docstrings.
- Saving and loading command-line configurations to/from json or yaml files for reproducibility.
- Building nested argument hierarchies where subcommands or modules each have their own configuration dataclass.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Transforms argparse scripts into structured, typed command-line interfaces using dataclasses, with support for nesting, inheritance, and automatic help generation from docstrings.

Yes. Low install friction, no known vulnerabilities, MIT license, and active maintenance make it a safe choice. It genuinely reduces argparse boilerplate for projects using dataclasses, especially those with reusable or nested argument groups. Suitable for both small scripts and larger CLI applications.

## Install

pip install simple-parsing
uv add simple-parsing
poetry add simple-parsing

## Installing simple-parsing

Before you install: Low friction: pure Python wheel with only two runtime dependencies (docstring-parser and typing-extensions). Actively maintained with recent release.

License in practice: MIT License permits unrestricted use, modification, and distribution with minimal obligations—suitable for any project type.

Quickstart:

from dataclasses import dataclass
from simple_parsing import ArgumentParser

@dataclass
class Options:
    log_dir: str
    learning_rate: float = 1e-4

parser = ArgumentParser()
parser.add_arguments(Options, dest="options")
args = parser.parse_args()
print(args.options)

Requires Python 3.9 or later.

Verify before relying:

- Whether nesting depth has practical limits or performance implications at scale.
- Whether serialization to json/yaml handles all dataclass field types automatically or requires custom handlers.
- Performance characteristics when parsing very large numbers of arguments or deeply nested structures.

## Package facts

- License: MIT License Copyright (c) 2019 Fabrice Normandin Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to… (full text in the JSON record) (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 4.3M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

typed argument parsing, dataclass command line arguments, argparse wrapper, structured CLI arguments, nested command line config, automatic help from docstrings, reusable argument groups, cli-framework, dataclass-driven

[View on SkillFed](https://skillfed.io/packages/simple-parsing) · [View on PyPI](https://pypi.org/project/simple-parsing/)