JonasGeiping/breaching

Breaching privacy in federated learning scenarios for vision and text

68
/ 100
Established

This framework helps data scientists and machine learning engineers evaluate privacy risks in federated learning systems. It takes in shared model updates from individual users and outputs the private training data that was used to generate those updates. This allows organizations deploying federated learning to understand how vulnerable their systems are to data reconstruction attacks and identify potential privacy breaches.

313 stars. Available on PyPI.

Use this if you are building or deploying federated learning models and need to assess their susceptibility to data privacy attacks.

Not ideal if you are looking for defenses or full end-to-end federated model training simulations, as this tool focuses purely on evaluating attack effectiveness.

federated-learning data-privacy model-security machine-learning-auditing privacy-engineering
Maintenance 10 / 25
Adoption 10 / 25
Maturity 25 / 25
Community 23 / 25

How are scores calculated?

Stars

313

Forks

73

Language

Python

License

MIT

Last pushed

Jan 24, 2026

Commits (30d)

0

Dependencies

5

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/JonasGeiping/breaching"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.