Skip to content

Axiomize

Reproducible scientific modeling that survives contact with reality.

Axiomize turns a vague idea into an explicit, versioned mathematical model, validates it dimensionally and numerically, and exports an artifact someone else can re-run years from now. Units are mandatory, constraints are named and justified, and expensive steps ask before they run.

Start here

The two ways to use it

As a Python library and CLI. pip install axiomize gives you the axiomize console entry point, the Model IR engine, and REST/MCP servers.

pip install axiomize
axiomize capabilities        # which backends are actually usable here
axiomize intake "your idea"  # clarify before committing to numbers

As an agent skill. Copy skills/axiomize/ into your agent's skills directory (~/.config/opencode/skills/ for opencode, ~/.claude/skills/ for Claude Code) and ask in plain language: "Model this idea mathematically: my gym keeps losing members after three months; how do I stop the churn?"

Both paths use the same engine and the same validation contract, so results do not depend on which surface you call.

What you get

  • Dimensional validation, so you cannot add meters to seconds
  • Scientific constraints as named, justified checks with a scientific basis
  • Numerical verification that separates discretization error from stochastic variability
  • Model-family comparison with recorded reasoning for the choice
  • Portable export: Model IR JSON, Python, YAML, notebooks, SBML, CellML, Modelica, GraphML
  • An integrity-checked run ledger for stored results

What it does not do

It does not make a weak model strong. It makes a weak model fail early and visibly. Bayesian sampling requires the full extra and FEM requires FEniCS/DOLFINx; both are reported as unavailable rather than silently replaced by something weaker.

SIR demo

Where to look next

If you want to Go to
Run a model in five minutes Quickstart
See finished problems per domain Example gallery
Know which lenses matter in your field Domain packs
Wire it into an agent (MCP, REST, CLI) Integrations
Decide what the benchmarks prove Benchmark results
Decide what to trust Security and SECURITY.md
Contribute a lens or example CONTRIBUTING.md