Installation

Install PyBaMM on Linux, macOS, or Windows using pip or conda.

PyBaMM is available on GNU/Linux, macOS, and Windows. It requires Python 3.10–3.14.

Install PyBaMM

pip

pip install pybamm

All required dependencies (NumPy, SciPy, CasADi, etc.) are installed automatically.

Set up a virtual environment

We strongly recommend installing PyBaMM inside a virtual environment so that it does not alter your system Python files.

1. Create a virtual environment

   Run the following command in the directory where you want to work:

   Linux and macOS

   virtualenv env
   
2. Activate the environment

   Linux and macOS

   source env/bin/activate
   
3. Install PyBaMM

   pip install pybamm
   
4. Deactivate when done

   deactivate
   

For a primer on Python virtual environments, see Real Python — Python Virtual Environments.

Optional dependencies

You can install PyBaMM with optional extras depending on your use case:

Extra	What it adds	Install command
plot	Matplotlib-based plotting via sim.plot()	pip install "pybamm[plot]"
cite	Citation management via pybamm.print_citations()	pip install "pybamm[cite]"
bpx	Battery Parameter eXchange format support	pip install "pybamm[bpx]"
jax	JAX-based solver (see restrictions below)	pip install "pybamm[jax]"
all	All of the above except JAX	pip install "pybamm[all]"

Optional — JAX solver

Install jax and jaxlib to use the JAX-based solver:

pip install "pybamm[jax]"

This automatically installs compatible versions of jax and jaxlib alongside PyBaMM.

The JAX solver requires Python 3.11 or later and is not supported on macOS with an x86_64 (Intel) processor. It works on Linux (x86_64 and ARM), macOS (Apple Silicon), and Windows.

The full conda-forge pybamm package includes jax and jaxlib by default.

Docker

A Dockerfile is available in the PyBaMM repository for containerised environments.

Uninstall

pip uninstall pybamm