Configuration Files in Python

configparser in the standard library reads and writes INI files and needs no dependencies. python-dotenv loads variables from a .env file into the environment, which is the usual way to configure local development without committing secrets. pydantic-settings declares settings as a typed Pydantic model and fills it from environment variables, .env files and secret files, with validation at startup so a missing or malformed value fails early.

Hydra, from Meta, composes hierarchical configuration from YAML files and lets you override any value from the command line, which is popular for machine learning experiments with many variants. Dynaconf layers settings from files, environment variables and external stores, supports per-environment sections such as development and production, and integrates with Django and Flask. For reading TOML, tomllib is in the standard library from Python 3.11.