NXP Partners with NVIDIA to Add TAO Toolkit to eIQ, Speeding Edge AI Development and Performance
Partnership sees TAO APIs added to eIQ as a one-stop platform for model training, customization, optimization, and deployment.
NXP Semiconductors has announced a partnership with NVIDIA to integrate the latter's TAO Toolkit application programming interfaces (APIs) with the former's eIQ machine learning development environment — promising speedier development of edge artificial intelligence (edge AI) applications.
"AI innovation will define the future of the smart connected world," claims NXP's Charles Dachs of the team-up between the two companies. "Combining NVIDIA's expertise in training and testing AI models with NXP's long history of industrial and IoT [Internet of Things] edge innovation creates a synergy that allows our customers to bring their AI models to market quickly and easily."
"NVIDIA TAO greatly simplifies the creation and deployment of AI models, including state-of-the-art generative AI models," adds Nvidia's Deepu Talla. "This collaboration brings highly accurate, optimized AI models tuned with NVIDIA TAO and plugged seamlessly into NXP's eIQ development environment to accelerate edge AI deployments."
NXP's eIQ platform is designed for the development and deployment of AI algorithms onto NXP's microcontrollers and microprocessors, and will now include direct integration with NVIDIA's TAO Toolkit APIs — delivering, the company claims, a single platform for model training, optimization, and deployment at the edge, with boosted performance as an added bonus.
The partnership promises to reduce development time through the use of a library of pre-trained models, which can then be customized for a particular task using transfer learning — then deployed onto compatible NXP devices, including those with neural processing unit (NPU) accelerators like the NXP i.MX 93 range.
More information on eIQ is available on NXP's website, with NVIDIA's TAO Toolkit documented on the company's own website.