---

CrypTorch is an MPC-based ML framework built on top of PyTorch, developed and maintained by the PAWS lab @Penn State. CrypTorch enables you to run your PyTorch-written ML program using multi-party computing (MPC), allowing model training and inference while ensuring the confidentiality of input data and model weights.

Description of the framework can be found in our arxiv paper.

Setting up CrypTorch

1. Clone this repo

   git clone https://github.com/psu-paws/CrypTorch.git
   cd CrypTorch
   

2. Setup a virtual environment (Optional)

   We recommend setting up and activating a virtual environment (venv, conda, etc) to avoid installing packages globally and potentially causing conflicts.

   • Conda (Example)

     conda create -n CrypTorch python=3.12
     conda activate CrypTorch
     

   • venv (Example)

     python3.12 -m venv .venv
     source .venv/bin/activate
     

3. Install dependencies

   1. Install PyTorch version 2.8.0 using instructions from PyTorch's website. For example, to install CPU-only version of PyTorch, use the following command. If you have GPUs, install a version of PyTorch 2.8.0 that is compatible with your cuda version instead.

      pip install torch==2.8.0 torchvision==0.23.0 torchaudio==2.8.0 --index-url https://download.pytorch.org/whl/cpu
      

   2. Install crypten-plus-plus (modified CrypTen)

      pip install -e ./CrypTen
      

   3. Install cryptorch

      pip install -e .
      

4. (When using GPU) Build GPU integer kernels Follow instructions under cutlass/README.md.

User Slack channel

To join our user Slack channel, use the following link.

Citing CrypTorch

To cite our work, use the below Bibtex:

@article{liu2025cryptorch,
  title={CrypTorch: PyTorch-based Auto-tuning Compiler for Machine Learning with Multi-party Computation},
  author={Liu, Jinyu and Tan, Gang and Maeng, Kiwan},
  journal={arXiv preprint arXiv:2511.19711},
  year={2025}
}

---

CrypTorch logo by @gold.and.glyphs.