Skip to content

Digital Twins that run simulations

MuPIF combines a semantic Digital Twin model with a distributed simulation platform. Describe your system once, then run any simulation workflow directly on its Digital Twin, with every result traceable to its origin.

Get started Partner with us

Workflows on the Digital Twin

One platform, three ideas

  • Semantic Digital Twin


    Describe your physical system as entities, attributes with units, and relations, defined by a JSON schema. MuPIF generates the database and keeps every state of every entity.

    Digital Twin

  • Distributed simulation platform


    Wrap existing codes, commercial or in-house, behind a common Model API. Connect them across institutions, servers and HPC clusters, securely over a VPN.

    Platform

  • Workflows on the Digital Twin


    Workflow inputs are read from the twin and outputs create new entity states. Every result is linked to the data and the workflow that produced it.

    How it works

How it works

Model your system

Define the entities of your Digital Twin and their relations in a JSON schema.

Plug in your models

Wrap simulation codes, data-driven or AI models with the MuPIF Model API.

Compose workflows

Write them in Python, or draw them in the low-code workflow editor.

Run & trace

Execute workflows on the twin. MuPIF DB records every input, output and state.


Core capabilities

  • Interoperability by design

    MuPIF standardizes operations on data, not data formats. Any existing code or data representation can be plugged in through standardized interfaces.

  • Distributed & secure

    Models run where they live and communicate transparently over a VPN or SSH. Partners keep their codes and data at home.

  • HPC integration

    HPC model servers submit jobs to batch systems, so heavy simulations run on HPC within the same workflow.

  • Traceability & reproducibility

    MuPIF DB keeps workflows, executions, inputs and outputs. Every result can be traced back and reproduced.

  • Units & metadata

    All quantities carry physical units. Models and workflows are described by schema-validated metadata.

  • Free & open source

    MuPIF is written in Python and released under the LGPLv3 licence, free for academic and commercial use.


Used in research and innovation projects

  • MuPIF provides the Digital Twin platform for materials diagnostics and monitoring of engineering infrastructure.

  • MuPIF is the multiscale modelling platform for organic and large area electronics, with a low-code workflow approach.

Also used in Composelector, MMP, DeeMa, SUMO and other projects. All use cases

Planning a new project?

We are looking for partners who need a Digital Twin or a coupled simulation platform. Partner with us and make MuPIF the platform of your next project.


Latest announcements

2026-10-22 New MuPIF website. The MuPIF pages have a new look and a clearer structure.

2025-04-22 Musicode low-code paper published. Streamlining multi-scale materials modeling: the Musicode low-code approach for simulation workflows and executable MODAs has been published in Integrating Materials and Manufacturing Innovation. Read the paper.

2024 Entity Data Model paper (open access). Towards digital twins: Design of an entity data model in the MuPIF simulation platform has been published in Advances in Engineering Software. Read the paper.

New Video tutorials. Learn the platform step by step, from the installation of the demo infrastructure to running workflows. Get started.


How to cite MuPIF

B. Patzák, S. Šulc and V. Šmilauer. Towards digital twins: Design of an entity data model in the MuPIF simulation platform. Advances in Engineering Software, 197, 2024. More publications


(since February, 2023)