OpenGradient Brings Decentralized AI to Sui with Walrus Integration for User-Owned Models
In a push to decentralize artificial intelligence, research lab OpenGradient has integrated Walrus, a decentralized data storage protocol, into the Sui Network. The move, announced on June 25, aims to give users and developers greater control over their AI models—making them more transparent, affordable, and private.
By leveraging Walrus, OpenGradient can now host over 100 AI models across various blockchain ecosystems. The goal is to replace centralized infrastructure—long dominated by tech giants—with a user-first approach to AI development. It’s part of a growing movement in the blockchain space to return ownership and agency to individuals in emerging technologies.
Decentralizing the Future of AI
“AI isn’t something that belongs to a handful of massive tech companies; it belongs to users and always has,” said Matthew Wang, co-founder and CEO of OpenGradient. “We want to make it possible for anyone—from experienced developers to everyday users—to create, modify, and own their own AI models.”
By moving AI to a decentralized architecture, OpenGradient hopes to remove barriers created by centralized data silos. This not only reduces costs but improves user privacy and provides visibility into how models are trained.
Walrus: Powering Privacy-Focused Storage
One of the key components of the integration is Walrus, a decentralized storage network designed to give users control over their data. Traditional AI development often relies on centralized cloud platforms, which come with trade-offs in privacy, cost, and transparency.
“The only options available until now have put control in the hands of centralized providers,” said Rebecca Simmonds, Managing Executive at the Walrus Foundation. “By integrating with OpenGradient, we’re giving developers and users a real alternative—one that keeps data in the hands of its rightful owners.”
Through Walrus, users can trace exactly what data was used to train a model, offering a level of transparency rarely seen in mainstream AI platforms.
Why It Matters
AI models rely on enormous datasets to function effectively. In a centralized system, that data is typically harvested and controlled by corporations. OpenGradient and Walrus propose a different model—one where anyone can contribute, inspect, and benefit from the data and models they interact with.
The integration on Sui, a high-performance Layer-1 blockchain, ensures these decentralized AI agents can scale across multiple environments, making it easier for developers to tap into a privacy-focused AI infrastructure.