Practical overview
When it fits
Use it when adapting a TensorFlow-oriented workflow to compute on encrypted data, or when evaluating MPC and homomorphic-encryption approaches for training or prediction.
Who it is for
- TensorFlow developers exploring computation over encrypted inputs.
- Teams or researchers evaluating privacy-preserving training or inference across parties.
- Developers prototyping encrypted machine-learning workflows with a Keras-like interface.
Documented uses
- Define private inputs and variables for an encrypted computation.
- Run TensorFlow-style operations, such as matrix multiplication, over encrypted data.
- Explore private model inference and training through the repository's examples and benchmarks.
Requirements & limits
- The documented PyPI package requires Python 3.8+ and TensorFlow 2.9.1+.
- TensorFlow 1 session and placeholder features are not supported by the TensorFlow 2-based version; the README points users needing that style to version 0.8.0.
- The README presents source installation as a basic setup and directs developers to further installation instructions for matching the PyPI package's performance and security.
- Published benchmarks use ABY3 across three machines and specify hardware and network conditions, so the figures are tied to that test setup.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Install TF Encrypted from PyPI
Set up the documented package prerequisites and install TF Encrypted.
Before you start
- Python 3.8 or later
- TensorFlow 2.9.1 or later
Steps
Install the required Python and TensorFlow versions.
Run the command below.
pip install tf-encrypted
Expected outcome
The README's PyPI installation route is followed for TF Encrypted.
Guide limitations
- These instructions cover package installation, not running a particular model or deployment.
- The README notes that a source installation may need extra steps to match the performance and security of the PyPI package.
Build TF Encrypted from source
Follow the repository's documented source installation steps for development or an uncompiled platform.
Before you start
- Python and TensorFlow environment suitable for the project
- Git and the
makebuild tool
Steps
Clone
https://github.com/tf-encrypted/tf-encrypted.git.Change into the
tf-encrypteddirectory.Run the command below.
pip install -e .Run the command below.
make build
Expected outcome
The README describes this as a basic source installation.
Guide limitations
- The README says additional installation steps are required to match the performance and security of the PyPI package.
- The guide does not cover executing a workload.
Compare options for Model inference and serving
These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.
| Project | Documented purpose | Platform | Access and requirements |
|---|---|---|---|
| Tf Encrypted | Define private inputs and variables for an encrypted computation. | Linux, macOS, Windows | |
| Hyperpose | Customize pose-estimation architectures and datasets. | Docker, Linux, macOS, Windows | |
| Multi Task Nlp | Define multiple tasks and their data and model settings in a YAML file. | Linux, macOS, Windows | The project is an open-source library under the Apache-2.0 license, distributed through its public GitHub repository. |
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 17 October 2026
- Original website title
- tf-encrypted/tf-encrypted · Official homepage title · retained in its source language