Ezkl lets you compile AI/ML models to zero-knowledge proofs
Appears in
35 ballots
Contribution
Ezkl is an open-source engine for doing inference for deep learning models and other computational graphs in a zk-snark (ZKML). It can be used from a Jupyter notebook, Python, Javascript, or the command line. It has brought a wave of new developers from Python, AI/ML, and data science because it lets them run the models they build with, effectively on-chain. And it all verifies on Optimism.
We've helped teach developers about ZK and build AI/ML models with verifiers that work on Optimism. Whether it is defi, on-chain games, identity, or something else, writing a few lines of Python and seeing the proof run and verify gives an aha experience that helps people get excited about what is possible on Optimism and on-chain in general.