Install Dyno¶

This page explains how to use Dyno in your own project. If you want to work on Dyno itself (clone the repository, run the tests, build the documentation), see Develop Dyno.

Dyno is available on PyPI and as conda packages from both conda-forge and the econforge channel on prefix.dev. Conda packages are also built for WebAssembly (WASM) and are available inside notebook.link. Any conda-compatible tool can install the conda packages, including conda and micromamba; we recommend Pixi for a reproducible project environment.

Install Pixi¶

If you do not have Pixi installed on your system:

Linux & macOSWindows (PowerShell)

curl -fsSL https://pixi.sh/install.sh | bash

iwr -useb https://pixi.sh/install.ps1 | iex

Add Dyno to a project¶

Create a project (skip this step to add Dyno to an existing Pixi project), add the econforge channel, then add the dyno package:

pixi init my-project
cd my-project
pixi workspace channel add --prepend https://prefix.dev/econforge
pixi add dyno

--prepend puts the econforge channel before conda-forge, so Pixi installs the latest Dyno release from econforge. Your pixi.toml then contains:

[workspace]
channels = ["https://prefix.dev/econforge", "conda-forge"]

[dependencies]
dyno = ">=0.1.14,<0.2"

(the exact version constraint depends on the current release).

If you prefer a different installer, the regular Python package can also be installed from PyPI:

python -m pip install dynopy

The optional components described below are distributed only as conda packages and are not available on PyPI. Install them with a conda-compatible tool such as Pixi, conda, or micromamba.

Optional components¶

Dyno Lab (JupyterLab extension). The jupyterlab-dyno package provides the extension; add JupyterLab alongside it, then launch it:

pixi add jupyterlab jupyterlab-dyno
pixi run jupyter lab

See the Dyno Lab documentation for details.

Dynare preprocessor. Dyno reads Dynare .mod files with its own parser, which needs no extra package. To use DynareModel, which relies on the official Dynare preprocessor, add the native dynare-preprocessor-pylib package:

pixi add dynare-preprocessor-pylib

See the Dynare quickstart for the differences between the two.

Verify the installation¶

pixi run python -c "from dyno import DynoModel; print('Dyno is installed')"

Then follow the Dyno quickstart to write and solve your first model.

Supported platforms¶

dyno and jupyterlab-dyno are pure-Python (noarch) packages and require Python 3.12 or 3.13, so they install wherever those Python versions are available from conda-forge. dynare-preprocessor-pylib contains compiled code and is only published for some platforms (currently linux-64, osx-64 and win-64; not osx-arm64).

Dyno is developed and tested on linux-64 only: the development workspace in the repository (platforms in pixi.toml) is restricted to that platform.