Connected nodes logo representing Federated Learning

Federated Learning Frameworks

Decentralized, privacy-preserving ML platforms

Federated Learning (FL) trains machine learning models across decentralized data sources — mobile devices, hospitals, or enterprise silos — without the raw data ever leaving its origin. Below are 11 actively maintained FL frameworks, each verified and linked to its current home.

01

FATE

An open-source project initiated by Webank's AI Department, providing a secure computing framework to support the federated AI ecosystem.

02

Substra

An open-source federated learning software providing a flexible Python library and web app to run FL training at scale, focused on healthcare and biotech.

Now a Linux Foundation Project
03

PySyft

An open-source platform enabling secure, privacy-preserving data science — remote execution, differential privacy, and encrypted computation for research.

Now OpenMined PySyft
04

TensorFlow Federated

An open-source framework for machine learning and other computations on decentralized data, developed to facilitate open research and experimentation.

05

IBM Federated Learning

A Python framework for federated learning in an enterprise environment, supporting DNNs, classic ML, and reinforcement learning with a large fusion-algorithm library.

Now on GitHub
06

NVIDIA FLARE

A domain-agnostic, open-source SDK for federated learning that adapts existing ML/DL workflows to a federated paradigm for secure, multi-party collaboration.

Now NVIDIA FLARE
07

Flower

A unified approach to federated learning, analytics, and evaluation — federate any workload, any ML framework, and any programming language.

Now Flower.ai
08

Fed-BioMed

An open-source platform for secure collaborative health data programs, letting healthcare data providers train AI models without moving patient data.

Now fedbiomed.org
09

FLUTE

Federated Learning Utilities for Testing and Experimentation — a platform for conducting high-performance federated learning simulations at scale.

Archived Project
10

FedLab

Provides the necessary modules for FL simulation, including communication, compression, model optimization, and data partitioning.

11

OpenFL

A Python library for federated learning that enables organizations to collaboratively train a model without sharing sensitive information.