brannondorsey/PassGAN
A GAN-based deep learning model for generating password candidates trained on leaked password datasets.
Not currently ranked — collecting fresh signals.
star history
PassGAN implements a Generative Adversarial Network architecture for password guessing, adapted from the Improved Training of Wasserstein GANs approach. The repository provides a command-line interface for training custom models on password datasets and sampling generated passwords from pretrained checkpoints. It includes a pretrained model trained on the RockYou dataset and supports training on additional leak datasets.
Frequently asked
- What is brannondorsey/PassGAN?
- A GAN-based deep learning model for generating password candidates trained on leaked password datasets.
- Is PassGAN open source?
- Yes — brannondorsey/PassGAN is open source, released under the MIT license.
- What language is PassGAN written in?
- brannondorsey/PassGAN is primarily written in Python.
- How popular is PassGAN?
- brannondorsey/PassGAN has 2k stars on GitHub.
- Where can I find PassGAN?
- brannondorsey/PassGAN is on GitHub at https://github.com/brannondorsey/PassGAN.