Unsupervised machine learning with multivariate Gaussian mixture model.
Open source
24 alternativesOpacus
A library that enables training PyTorch models with differential privacy.
About Opacus
Opacus is an open-source library for training PyTorch models with differential privacy. Users create a PrivacyEngine and pass their model, optimizer, and data loader to its make_private() method with settings such as noise multiplier and maximum gradient norm.
It includes tracking of the privacy budget expended during training, support for Fast Gradient Clipping and Ghost Clipping, and examples and tutorials covering MNIST, BERT text classification, image classification, and LSTM name classification.
Key features
- Train PyTorch models with differential privacy via PrivacyEngine
- Convert model, optimizer, and data loader using make_private()
- Track privacy budget expended during training
- Support Fast Gradient Clipping and Ghost Clipping
- Compatible with LoRA and peft library for DP-SGD training
- Include examples for MNIST, BERT, image classifier, and LSTM
Best for: ML practitioners and differential privacy researchers who train PyTorch models with differential privacy.
Good for
- Requires passing model, optimizer, and data loader to make_private()
- Allows tracking of privacy budget expended during training
- Provides examples and IPython-based tutorials for common model types
Watch out for
- Installing directly from source may include quirks and occasional bugs
Opacus at a glance
- Website
- github.com/pytorch/opacus
- Category
- AI & Machine Learning › Models & ML Research
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
Opacus alternatives
Tools and sites you could use instead, ranked by how closely they match.
Alibi explains machine learning model predictions through a set of algorithms.
BigARTM implements topic modeling based on additive regularization, with documentation and downloads available.
timm provides PyTorch image encoders and backbones along with training, evaluation, inference, and export scripts.
Open source
24 alternatives
Code and documentation to train Stanford's Alpaca models, and generate the data.
Open source
24 alternatives
The learn2learn library supports meta learning experiments and research with PyTorch.
Open source
24 alternativesPopular in Models & ML Research
Well-known listings from the same topic.
Falcon LLMfalconllm.tii.aeERNIEernie.baidu.comTokenVersetoken-verse.github.iowTAObittensor.comAlpacacrfm.stanford.edu/2023/03/13/alpaca.htmlSmolVLAhuggingface.co/blog/smolvla
Opacus FAQ
How do I install Opacus?
Install the latest release with pip install opacus or conda install -c conda-forge opacus.
How do I make a PyTorch training run private?
Create a PrivacyEngine and pass the model, optimizer, and data loader to make_private() with noise_multiplier and max_grad_norm settings.
Can Opacus be used with LoRA?
A tutorial shows how the LoRA and peft library can be used with DP-SGD.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.