pyserde

What it is and what it does

pyserde is a serialization library built on top of Python dataclasses that converts objects to and from multiple formats (JSON, YAML, TOML, MsgPack, Pickle, dict) using a decorator. You annotate your dataclass fields with PEP 484 type hints, and the library generates serialization and deserialization code automatically. It handles a wide range of types including primitives, standard containers (list, dict, set, tuple), Optional and Union types, datetime objects, UUID, Path, Decimal, Enum, and numpy types.

The library focuses on simplicity: declare your class, decorate it, and call to_json() or from_json() to convert. It supports field-level customization (rename, alias, skip conditions, custom serializers), class-level attributes (case conversion, custom serializers), and advanced features like flattening, forward references, and generic types. Dependencies are minimal and well-established (beartype for runtime type checking, jinja2 for code generation, plum-dispatch for method dispatch).

Use it for:

• Convert API request/response dataclasses to JSON for REST endpoints or webhooks.
• Load configuration files (YAML, TOML) into typed dataclass objects with validation.
• Serialize domain objects to multiple formats (JSON for APIs, Pickle for caching) from a single definition.
• Build data pipelines where objects move between services in different formats without manual conversion.
• Validate and transform incoming JSON data into strongly-typed Python objects.
• Store and retrieve structured data with automatic type checking during deserialization.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Serialize and deserialize Python dataclasses to and from JSON, YAML, TOML, MsgPack, Pickle, and dict formats using a decorator-based API.

Yes. pyserde is actively maintained, has no known vulnerabilities, supports current Python versions (3.10–3.14), and offers a clean, decorator-based API for multi-format serialization of dataclasses. Install friction is low and the dependency set is reasonable. It is a solid choice for projects that need type-safe serialization without the overhead of heavier frameworks.

Install

pyserde on PyPI

pip install pyserde

uv add pyserde

poetry add pyserde

Installing pyserde

Before you install

Low friction install with a pure-Python wheel. Actively maintained with a release 4 days ago; repository shows 851 stars and recent activity. Six runtime dependencies (beartype, casefy, jinja2, plum-dispatch, typing-extensions, typing-inspect) are all standard utilities with minimal overhead.

License in practice

MIT License permits unrestricted use, modification, and distribution in both open-source and commercial projects with only attribution required.

Quickstart

pip install pyserde

from pyserde import to_json, from_json

@serde
class Foo:
    i: int
    s: str

obj = Foo(i=10, s='foo')
json_str = to_json(obj)
obj_restored = from_json(Foo, json_str)

Requires Python 3.10 or later (3.10, 3.11, 3.12, 3.13, 3.14 supported).

Verify before relying

• Performance characteristics compared to other serialization libraries in typical workloads.
• Maturity and stability of experimental SQLAlchemy integration.
• Whether custom (de)serializers handle all edge cases users encounter in production.

Package facts

Evidence: pyserde-0.32.0-py3-none-any.whl

Tags

More Software Development packages