SoundHound AI (Nasdaq: SOUN) announced OASYS Edge, an embedded architecture designed to run large language model-powered voice AI agents directly on edge hardware, without requiring a cloud connection. The product is intended for vehicles and smart devices and is scheduled for deployment in late 2026.

According to the company, OASYS Edge allows developers to build a voice AI agent once on the OASYS platform and then deploy it in the cloud, on the edge, or in a hybrid environment. The company claims this makes it the first platform enabling global automakers, device manufacturers, and service providers to deploy fully embedded agentic AI.

Because processing occurs locally on the device, voice interactions and user data are not transmitted to external servers by default. SoundHound says this approach also reduces recurring cloud API costs and bandwidth demands for manufacturers.

The system is designed to handle multi-step tasks without network access — such as route planning with points of interest — and to support continuously listening devices like robots without streaming audio to the cloud.

"For a long time, the industry assumed this kind of sophisticated agentic capability could only live in the cloud, forcing companies to compromise on speed, cost, reliability, and privacy," said James Hom, Chief Product Officer and Co-Founder at SoundHound AI. "OASYS Edge fundamentally shifts that dynamic."

Live demonstrations of OASYS Edge are currently available, and the company plans to feature the product at CES 2027.