Bionetta: Efficient Client-Side Zero-Knowledge Machine Learning Proving

2025-10-08

AI Summary

Deploying zero-knowledge machine learning (ZKML) on resource-constrained clients (e.g., mobile devices) and native EVM smart contracts remains challenging due to prohibitive computational and on-chain verification overhead. Method: This paper introduces the first end-to-end, EVM-deployable lightweight ZKML framework. Leveraging the UltraGroth protocol, it designs customized neural network circuits, one-time preprocessing, and an efficient trusted setup generation mechanism—substantially reducing proof size and verification cost. Contribution/Results: The framework enables full inference and proof generation directly on mobile devices, with proof generation time orders of magnitude faster than state-of-the-art tools. Generated proofs are efficiently verifiable on-chain, requiring less than 1M gas. Crucially, it achieves, for the first time, native EVM compatibility and practical mobile deployment within a single ZKML system—paving the way for privacy-preserving, on-chain AI verification.

Technology Category

• Machine Learning: Privacy
• Natural Language Processing: (Large) Language Models
• Computer Vision: Large Vision Models

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Security and Privacy: Security and privacy of machine learning and AI applications

Abstract

In this report, we compare the performance of our UltraGroth-based zero-knowledge machine learning framework Bionetta to other tools of similar purpose such as EZKL, Lagrange's deep-prove, or zkml. The results show a significant boost in the proving time for custom-crafted neural networks: they can be proven even on mobile devices, enabling numerous client-side proving applications. While our scheme increases the cost of one-time preprocessing steps, such as circuit compilation and generating trusted setup, our approach is, to the best of our knowledge, the only one that is deployable on the native EVM smart contracts without overwhelming proof size and verification overheads.

Problem

• Enables client-side zero-knowledge machine learning proofs
• Optimizes proving time for neural networks on mobile devices
• Achieves native EVM compatibility with practical verification overheads

Innovation

• UltraGroth-based zero-knowledge machine learning framework
• Enables client-side proving on mobile devices
• Deployable on native EVM smart contracts