I'm LongbridgeAI, I can summarize articles.Sunrun launched a pilot program to transform solar-powered homes into distributed AI data centers. Homeowners with Sunrun solar and battery systems can host AI compute nodes and earn payments for providing computing capacity. This initiative allows Sunrun to sell inference workloads to enterprise customers, leveraging its large customer base to reduce latency compared to traditional centralized data centers.
Sunrun wants to turn thousands of solar-powered homes into a giant distributed AI data center – and pay homeowners to help power the AI boom.
The residential solar and storage company today launched a pilot program that installs AI compute nodes – that’s a server in a distributed system like the one pictured above – in homes equipped with Sunrun solar panels and battery storage. Homeowners who participate in the pilot will be paid for hosting the hardware, and Sunrun will sell the computing capacity to enterprise customers.
It sees Sunrun expanding beyond home renewable energy and virtual power plants into AI infrastructure. Instead of building big data centers, the company wants to use its network of more than 1.1 million existing customers to process AI inference workloads.
Inference is the stage where trained AI models generate responses to users. Unlike AI training, which requires massive data centers packed with graphics processing units (GPUs), inference workloads can often be spread across many smaller locations and benefit from being closer to end users, reducing latency.