Asynmov

Generate large corpora of synthetic data for world building. Asynmov pairs a Python API with a high-performance Rust core so you can produce millions of coherent entities, events, and relationships — the raw material for fiction, games, simulation, and LLM training data — without hitting a performance wall.

Features

• Entity generation — characters with internally consistent attributes driven by world configuration
• Event generation — causally ordered life events (birth, migration, occupation, death, and more)
• Relationship graphs — social, family, and professional networks across the population
• Configuration-first — describe the world in a TOML file; Asynmov handles the rest
• Deterministic — supply a seed and get identical output every time
• Multiple output formats — JSON, JSONL, and CSV, ready to load into any downstream tool
• CLI included — asynmov generate and asynmov validate for scripted pipelines

Installation

pip install asynmov

Pre-built wheels are provided for Linux (x86_64, aarch64), macOS (x86_64, arm64), and Windows (x86_64). No Rust toolchain required to use the package.

Quick start

from asynmov import World

world = World.from_config("my_world.toml")
corpus = world.generate(scale=1000)
corpus.export("output/", format="json")

A minimal config:

[world]
name = "The Gilded Republic"
era = "1880-1920"
setting = "american_historical"
seed = 42

[population]
size = 1000
regions = ["northeast", "midwest", "south"]

CLI

# Generate a corpus
asynmov generate --config my_world.toml --scale 5000 --output output/ --format jsonl

# Validate a config without generating
asynmov validate my_world.toml

Development

Requires a Rust toolchain (rustup.rs) and maturin.

pip install maturin
maturin develop      # compile Rust extension and install into current venv
pytest               # run tests

Documentation

Full design and architecture documentation lives in docs/.

License

MIT — see LICENSE.