Surrey unveils Move LLM for blockchain development

This groundbreaking release makes it faster, safer, and easier for Move Language developers to build intelligent Web3 applications, bridging the gap between smart contracts and artificial intelligence.

Endless Move-LM can understand, generate, and optimize Move smart contract code, enabling developers to create secure and efficient applications at unprecedented speed. Developed jointly by the University of Surrey’s Blockchain and Metaverse Application Academy and Endless Protocol, the model is fully open-source and available today on GitHub and the Endless Developer Platform. 

“Move represents the next generation of Web3 smart contract languages, and integrating AI unlocks its full potential,” said Professor Yu Xiong FAcSS, Director of SABMA at the University of Surrey. “With Endless Move-LM, developers gain an interpretable, trainable, and scalable tool that brings true AI to Move smart contract development.”

Endless Protocol, a decentralized platform built on the Move language, leverages its Move Virtual Machine (MoveVM) and modular framework to enable sophisticated, resource-secure applications. With Move-LM, developers can generate Move modules using natural language, detect resource-related errors automatically, and receive real-time security and optimization recommendations directly within the Endless Web3 environment.

“Move’s resource model ensures asset security but also creates complexity for developers,” added Xiong. “Endless Move-LM breaks down these barriers, letting teams focus on innovation while maintaining the highest security standards.”

Powering Intelligent Web3 Development

Endless Move-LM integrates tightly with Endless Chain, a next-generation Web3 infrastructure for intelligent component workflows, allowing developers to:

• Generate Move modules using natural language
• Automatically detect resource-related errors
• Receive real-time security and optimization guidance
• Build AI-assisted applications directly within the Endless Web3 ecosystem

“Move’s strict resource model is a boon for security but raises the technical threshold,” said Esra Ozturk, Head of Product at Luffa. “Endless Move-LM lowers that barrier, helping developers build blockchain applications more intuitively and fostering a truly AI-native Web3 ecosystem.”

The launch of Endless Move-LM positions the University of Surrey and Endless Protocol at the forefront of AI-driven smart contract research. Future exploration will focus on multi-chain compatibility, automated auditing, and autonomous AI agents, making Move smart contracts more accessible, intelligent, and secure than ever. 

Availability

Endless Move-LM is fully open-source and available today on GitHub and via the Endless Developer Platform.