konradhalas/dacite

Health v2 · maintenance only

Adoption not part of the score

2063 stars · 119 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Dacite is a lightweight Python library that simplifies creating data class instances from dictionaries, with support for nested structures, optionals, unions, generics, and custom type hooks. It focuses purely on deserialization into typed data classes, not data validation.

Use cases

• convert a dictionary into a typed dataclass instance
• deserialize HTTP request payloads into data transfer objects
• map raw database rows to data classes
• build nested dataclass structures from JSON-like dicts
• create type-hinted DTOs from untyped dict data

When to choose

• you use Python dataclasses and need to instantiate them from dictionaries
• you want lightweight deserialization without heavyweight validation schemas
• you need support for nested structures, unions, and generics in dict-to-dataclass conversion

When to avoid

• you need full input validation - use pydantic or marshmallow instead
• you need serialization back to dictionaries or JSON
• you are not using Python 3.7+ or dataclasses

Facets

library · maturity stable

serialization parser developer-tools backend python cross-platform dataclasses dto type-hints deserialization from-dict data-engineering

Member repositories

For agents

markdown · JSON · MCP: product_card(name="konradhalas/dacite")