OpenMined/PySyft

Health v2 · maintenance only

Adoption not part of the score

9957 stars · 1999 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

PySyft is a Python library that lets data scientists run computations on private data that stays on the data owner's server, with results shared only after owner approval. It is part of the OpenMined ecosystem for privacy-preserving, federated, and secure multi-party computation.

Use cases

• run data science on private data without moving it
• federated learning across data silos
• secure multi-party computation in python
• analyze sensitive medical or genomic data remotely
• let auditors query models without exposing weights
• collaborate on datasets under access control

When to choose

• you need to compute on data you cannot or should not copy
• data owners must approve every analysis before results are shared
• you want privacy-preserving machine learning with PyTorch

When to avoid

• you just need standard ML on data you already own
• you need the legacy PySyft ≤0.9 API without pinning versions
• you require a turnkey enterprise product rather than an open-source framework

Facets

library · maturity active

machine-learning deep-learning cryptography security privacy data-science sdk privacy machine-learning data-science security healthcare large-language-models python cross-platform cloud federated-learning secure-computation privacy-preserving remote-data-science data-sovereignty pytorch openmined

Member repositories

For agents

markdown · JSON · MCP: product_card(name="OpenMined/PySyft")