View all newsletters
Receive our newsletter - data, insights and analysis delivered to you
  1. Research Reports
April 25, 2022updated 27 Jul 2022 5:27pm

Syntiant Introduces AI Chip for Low-Power Edge AI Deployment

Concept: California’s tech startup Syntiant has introduced an AI chip for deep learning called the Syntiant NDP200 processor, which is suitable for always-on applications in battery-powered devices. Deep learning and semiconductor design are combined in this chip solution to create ultra-low-power, high-performance deep neural network processors. Syntiant’s silicon architecture, in conjunction with an edge-optimized training pipeline and data platform, aims to make edge AI deployments possible worldwide.

Free Whitepaper
img

Out of the Blue: Dynamic Liquid Crystal Windows

From cutting down on carbon emissions and maximising energy efficiency to enhancing the health and well-being of a buildings’ occupants, the latest ‘smart glass’ and ‘dynamic glazing’ solutions offer unparalleled opportunities. Designed to meet, if not exceed, standards for low and zero-emission buildings, Dynamic Liquid Crystal Windows by Eyrise provide the opportunity to minimise the need for artificial heating and cooling by optimising the use of that generated naturally. In this latest report, the technology’s performance is compared against its closest competitors on colour-rendering, switching speed, and circadian well-being to show where it leads the way. Gaze into the future of glazing by accessing this free whitepaper below.
by Merck
Enter your details here to receive your free Whitepaper.

Nature of Disruption: The NDP200 uses neural processing to run numerous programs at the same time while consuming minimum battery power. It is built on the Syntiant Core 2 programmable deep learning architecture and is designed to execute deep neural networks (DNN) natively on a range of topologies, including CNN (Convolution Neural Networks), RNN (Recurrent Neural Networks), and fully connected networks. At less than 1mW, it does visual processing with highly accurate inference. The NDP200 provides 25 times the tensor throughput of the Syntiant Core 1 found in the Syntiant NDP100, which is presently shipping in large quantities. For feature extraction and signal processing, a programmable Tensilica Hifi3 DSP is included. Edge devices can hear, speak, see, and feel owing to the startup’s Neural Decision Processors (NDP). Compared to current low-power MCU-based solutions, Syntiant NDPs typically offer a100x efficiency improvement while providing a greater than 10x increase in throughput. This allows larger networks at significantly lower power and greatly increases battery life while reducing total infrastructure costs.

Outlook: Syntiant aims to enable device intelligence by introducing orders of magnitude more neural computing than what was previously achievable. Earbuds, wearables, smartphones, smart speakers, laptops, vehicles, and other IoT consumer and industrial use cases can all benefit from the solution. Syntiant has raised $55M in new funding in March 2022 led by Millennium Technology Value Partners, Renesas Electronics Corporation, and Mirae Asset Capital. The fresh funding can go toward improving software capabilities and speeding up the deployment of Syntiant’s third-generation Syntiant Core 3 architecture, which the startup plans to launch in 2023. Microsoft’s M12, Intel Capital, Applied Ventures, the Amazon Alexa Fund, Robert Bosch Venture Capital, and Atlantic Bridge Capital are among the strategic and financial investors of Syntiant.

This article was originally published in Verdict.co.uk

Related Companies

Free Whitepaper
img

Out of the Blue: Dynamic Liquid Crystal Windows

From cutting down on carbon emissions and maximising energy efficiency to enhancing the health and well-being of a buildings’ occupants, the latest ‘smart glass’ and ‘dynamic glazing’ solutions offer unparalleled opportunities. Designed to meet, if not exceed, standards for low and zero-emission buildings, Dynamic Liquid Crystal Windows by Eyrise provide the opportunity to minimise the need for artificial heating and cooling by optimising the use of that generated naturally. In this latest report, the technology’s performance is compared against its closest competitors on colour-rendering, switching speed, and circadian well-being to show where it leads the way. Gaze into the future of glazing by accessing this free whitepaper below.
by Merck
Enter your details here to receive your free Whitepaper.

NEWSLETTER Sign up Tick the boxes of the newsletters you would like to receive.
I consent to GlobalData UK Limited collecting my details provided via this form in accordance with the Privacy Policy
SUBSCRIBED

THANK YOU

Thank you for subscribing to Design Build Network