Axelar unveiled AgentFlux, an open source framework designed to run AI agents locally while keeping private keys, trading strategies and customer data out of the cloud – a pitch aimed squarely at institutions exploring onchain finance but wary of privacy risks.
Developed by Interop Labs, the team behind the Axelar network, AgentFlux allows financial firms to deploy “agentic” automation without sending sensitive information to external infrastructure, the company announced Thursday.
The framework addresses one of the biggest frictions in AI-driven crypto operations: tool-calling.
Most agents today rely on cloud-based models to decide which blockchain tools to use and how to structure transactions, which can inadvertently expose the very information institutions want to protect. AgentFlux breaks these tasks into two smaller, specialized models: one for choosing the right tool and another for generating the arguments to be executed. According to the team behind Axelar, this setup improves tool invocation accuracy by 46% in benchmark tests, bringing local models closer to the performance of larger cloud systems.
Sergey Gorbunov, the co-founder of Axelar, shared in an interview with CoinDesk that he sees AgentFlux benefiting multiple tasks: “First, there are very sophisticated trading strategists, they are incredibly proprietary,” he said. “You never want to make sure that public cloud models are as good as executing your strategy the way you do. And when you upload them, you inherently enjoy violating them. So that’s one.”
“The second one I see is that blockchains are also widely used for analytics. For example, you’re doing tax reporting or investigating who made what transactions. Or maybe you have a specific fingerprint and you expect malicious activity on a few accounts. Maybe you want to upload that fingerprint to a cloud model or an AI system, and ask it to find out all the correlated transactions and all this suspicious behavior and patterns from this account. So you can completely keep that kind of activity. And do it privately,” Gorbunov added.
AgentFlux also fits in with Axelar’s broader multichain strategy. The team positions Axelar as a “gateway to onchain finance,” providing infrastructure for institutions to move assets and data across blockchains from a single point of integration. AgentFlux could enable a single AI agent to view risk, assess exposure, and execute trades across multiple ecosystems – a capability the company sees as essential for institutional adoption.
Read more: Canary Capital Files Launches ETF Tracking Cross-Chain Protocol Axelar