zkonduit/ezkl
A Rust library and CLI for proving deep learning model inference using zk-SNARKs.
Not currently ranked — collecting fresh signals.
star history
ezkl converts neural networks defined in PyTorch or TensorFlow into ONNX format, then generates ZK-SNARK circuits that can prove computation was performed correctly without revealing model weights or input data. It enables privacy-preserving ML inference where users can demonstrate that a model produced a specific output on specific inputs, without exposing either. The system uses the Halo2 proof system backend and supports verification on Ethereum or in browsers.
Frequently asked
- What is zkonduit/ezkl?
- A Rust library and CLI for proving deep learning model inference using zk-SNARKs.
- Is ezkl open source?
- Yes — zkonduit/ezkl is an open-source project tracked on heatdrop.
- What language is ezkl written in?
- zkonduit/ezkl is primarily written in Rust.
- How popular is ezkl?
- zkonduit/ezkl has 1.2k stars on GitHub.
- Where can I find ezkl?
- zkonduit/ezkl is on GitHub at https://github.com/zkonduit/ezkl.