# secretflow/secretflow

A unified framework for privacy-preserving data analysis and machine learning

Repository: https://github.com/secretflow/secretflow
Canonical: https://ross.abutalabs.com/products/secretflow
Homepage: https://www.secretflow.org.cn/docs/secretflow/en/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: differential-privacy, homomorphic-encryption, machine-learning, privacy-preserving, private-set-intersection, secure-multiparty-computation, trusted-execution-environment, data-analysis, federated-learning, split-learning, confidential-computing
Last push: 2026-04-24T07:57:55+00:00

## Health v2 (maintenance only)
Score: 68/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 79, release rhythm 37, longevity 100
- inputs: {"age_days": 1601, "days_push": 131, "days_rel": 341, "gap_med": 72.0, "n_releases_24m": 7}
- flags: prerelease_only
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2695, forks 470 (observed 2026-08-28T04:07:11.170068+00:00)

## What it is
SecretFlow is a unified Python framework for privacy-preserving data analysis and machine learning. It layers cryptographic devices (MPC, HE, TEE), a device-flow DAG abstraction, and algorithm/workflow layers for training on horizontally or vertically partitioned data.

## Use cases
- train machine learning models on federated data without sharing raw data
- run joint data analysis across parties with secure multiparty computation
- apply differential privacy to model training
- compute private set intersection between two datasets
- do vertical federated learning on feature-partitioned data
- build privacy-preserving data pipelines with hyperparameter tuning

## When to choose
- you need multi-party analytics or ML where data cannot leave each party
- you want a single framework covering MPC, HE, TEE, and federated learning
- you need both horizontal and vertical data partitioning support

## When to avoid
- you only need simple single-party ML without privacy constraints
- you need a lightweight production deployment without container orchestration
- your team cannot handle the operational complexity of cryptographic protocols

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, cryptography, data-science, security, privacy, rag
- domain: privacy, machine-learning, data-science, security, artificial-intelligence
- platform: python
- tags: privacy-preserving, federated-learning, secure-multiparty-computation, homomorphic-encryption, differential-privacy, private-set-intersection, trusted-execution-environment, split-learning, confidential-computing, linux, docker

## Member repositories
- secretflow/secretflow (main) score 68

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:11.170068+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:16:20.219318+00:00, confidence not recorded.
  - readme: https://github.com/secretflow/secretflow (fetched 2026-08-28T04:07:11.170068+00:00, sha 0b78210ddf8a)
  - homepage: https://www.secretflow.org.cn/docs/secretflow/en/ (fetched 2026-08-29T09:59:35.854256+00:00, sha 9ccc1111f64c)
  - registry_pypi: https://pypi.org/pypi/secretflow/json (fetched 2026-08-29T09:59:35.856533+00:00, sha 1ed620bc3323)
- Data as of 2026-08-30T08:39:29.467469+00:00.