MPyC: Multiparty Computation in Python

See MPyC--Python Package for Secure Multiparty Computation (PDF Slides) for some background information.

Use pip install mpyc for version 0.10 (April 16, 2024) of MPyC on PyPI. The documentation of this version is available at MPyC on github.io.

Use pip install git+https://github.com/lschoe/mpyc to get the latest version from GitHub, with up-to-date documentation at MPyC on readthedocs.io.

See github.com/lschoe/mpyc for source code and demos (Python scripts as well as Jupyter notebooks).

See what others are doing with MPyC: scholar.google.com/scholar?q=mpyc+multiparty.

hMPC for Multiparty Computation in Haskell!

The hMPC package implements a nontrivial subset of MPyC, covering support for the id3gini.py demo for privacy-preserving training of ID3 decision trees, see Id3gini.hs.

Run MPyC in your browser!

• Basic tests running a few MPyC demos with one party only: MPyC in PyScript and MPyC REPL from PyScript.
• For the real thing, try out Emil Nikolov's amazing MPyC Web:
  1. Click MPyC Web demo to run any MPyC demo on your PC, laptop, tablet, or phone with one party.
  2. Open MPyC Web demo in more places (parties on the same device or on other devices).
  3. Connect the parties after copy/pasting or sharing the provided ID.
     You can also scan the QR code to open the demo and copy the provided ID in one go.
  4. Run any MPyC demo between all succesfully connected parties!

Try it out in the cloud!

• Click to try the notebook SecureSortingNetsExplained.ipynb.
• Click to try the notebook SecureSantaExplained.ipynb, including example runs with multiple parties.
• Click to try the notebook OneWayHashChainsExplained.ipynb.
• Click to try the notebook KaplanMeierSurvivalExplained.ipynb.

Also, preview future extension to "Verifiable MPyC", see github.com/toonsegers/verifiable_mpc for code and ZKProof 2019 video (ZKProof 2019 PowerPoint) for a presentation.