Ceva and embedUR launch ModelNova to accelerate edge AI on NeuPro NPUs

The platform comes with pre-built models, datasets and use-case blueprints thus accelerating the time to prototype and deploy AI features.

“ModelNova is a powerful extension of our NeuPro ecosystem, giving developers access to a diverse and growing library of optimized AI models,” says Elia Shenberger, senior director of product marketing, AI Division at Ceva. “Our partnership with embedUR enables customers to build smarter, more responsive edge devices with minimal development overhead.”

The models are deployed and optimized using TensorFlow Lite Micro, which is integrated with Ceva’s NeuPro Studio SDK, enabling applications such as object detection, face recognition and anomaly detection.

The initiative focuses on “physical AI,“ which basically injects intelligence right into the device to make it operate autonomously without so much dependence on cloud connectivity.

Ceva’s highly scalable NeuPro NPU family is tailored to meet the ultra-low power and high-performance requirements in a broad range of edge AI applications, from entry-level to high-end connected devices.

The ModelNova offering is immediately available and will add new models and use cases on a rolling basis.

Earlier this year, Ceva announced a partnership with ALi Corporation integrating Ceva’s NPUs into ALi’s Video Display Sub-System (VDSS) platform. 

Article Topics

AI/ML  |  Ceva  |  device intelligence  |  edge AI  |  edge computing  |  embedded systems  |  embedUR  |  NPU