At Computex 2024, MediaTek announced its integration of Nvidia’s TAO platform with its own NeuroPilot SDK. MediaTek wnats to enhance its edge (artificial intelligence) AI capabilities in (Internet of Things) IoT applications. 

The partnership between the chipmarkers, Nvidia and MediaTek, aims to simplify the development process for companies and developers looking to leverage AI technologies across various sectors, including smart retail, healthcare, transportation, and more.

MediaTek, known for powering over two billion connected devices annually, has focused on optimizing its edge silicon portfolio to support efficient and powerful AI applications. By integrating Nvidia TAO, MediaTek’s NeuroPilot SDK will provide developers with a robust platform to create advanced AI solutions across a broad range of devices.

“Integrating Nvidia TAO with MediaTek NeuroPilot will further expand our vision of democratizing access to AI, helping drive a new wave of AI-powered devices and experiences,” said CK Wang, General Manager of MediaTek’s IoT business unit. 

Wang highlighted the ease of designing innovative edge AI products with MediaTek’s Genio line, which now benefits from enhanced AI capabilities through this integration.

Generative AI

Nvidia TAO offers over 100 ready-to-use pre-trained models, which will be accessible through MediaTek’s NeuroPilot SDK. This integration simplifies the process of tuning and deploying AI models, significantly reducing development time and complexity. Developers, even those without extensive AI expertise, can rapidly bring products to market, equipped with customized vision AI capabilities.

“Generative AI is turbocharging computer vision with higher accuracy for AIoT and edge applications,” said Deepu Talla, VP of Robotics and Edge Computing, Nvidia. 

Talla added that the combined strength of Nvidia TAO and MediaTek NeuroPilot SDK will bring advanced vision AI models to billions of IoT devices, enabling a new era of AI-driven solutions.

Discover more from Back End News

Subscribe now to keep reading and get access to the full archive.

Continue reading