AlphaICs begins global sampling of ‘Gluon

0

image

announces marketing partnership with CBC Co in Japan

AlphaICs introduces Gluon chip and Gluon board

AlphaICs introduces Gluon chip and Gluon board

AlphaICs introduces Gluon chip and Gluon board

MILPITAS, Calif., Feb. 13, 2022 (GLOBE NEWSWIRE) — AlphaICs, a leading AI fabless semiconductor company developing edge inference and edge learning technologies, has announced the availability of technical samples of “gluon’ – an 8 TOPS Edge AI inference coprocessor for surveillance, industrial, retail, auto and industrial IoT customers, offering best-in-class FPS/Watt performance.

Gluon comes with a complete (Software Development Kit) SDK that allows easy deployment of neural networks. The advanced edge inference chip offers customers the ability to inject AI capabilities into the current X86/ARM-based systems, resulting in significant cost savings.

Gluon offers the best fps/watt performance on the market for neural network classification and detection – 32 frames per second (FPS)/watt for Yolo-V2, a leading object detection model & 22 frames per second (FPS)/watt for VGG – 19, a leading classification model.

Gluon is currently being sampled for early customers to enable development of their imaging applications. It is designed for OEMs and solution providers targeting vision market segments such as surveillance, industrial, retail, industrial IoT and edge gateway manufacturers.

To accelerate its advance into the highly demanding silicon markets, AlphaICs has established a reseller relationship with CBC Co. Ltd, a Japanese company offering video surveillance products to its customers.

“CBC has been working with AlphaICs for almost two years and we are delighted to be their marketing partner in Japan. Introduced at the Japan AI Expo in October 2021, Gluon has attracted great interest in machine vision applications from Japanese customers due to its superior performance. AlphaIC’s co-processor strategy has been well received and we are very pleased to be able to offer this technology to our customers,” said Kazuhiko Kondo, Executive Officer, CBC Co., Ltd.

AlphaIC CEO Pradeep Vajram said “We are pleased with the results of our gluon silicon and are now demonstrating the innovative technology to our customers. Our team has worked very hard to create this high performance deep learning coprocessor that is gaining traction in the industry. Gluon is future-ready and well positioned to address the AI ​​vision applications for surveillance, retail, industrial and smart city markets.”

Early last year, the company raised $8 million to advance the development of Gluon based on the proprietary RAP architectureTM. AlphaICs highly scalable and modular architecture uses a specialized instruction set architecture specifically optimized for AI.

How is Gluon different from the current generation of edge inference chips?

Gluon, an 8 TOPS Edge AI coprocessor, is manufactured in 16nm FinFET process at TSMC. This chip accelerates deep learning neural network models for classification, detection, and segmentation, and is focused on vision applications. Gluon includes PCIe and LPDDR4 interfaces to enable high-speed transfers to host processors and DRAM, respectively.

Based on many innovations, Gluon offers first class benchmarks:

  • 153 frames per second with Yolo-V2, a leading object detection model (image size 416 x 416 x 3) at 4.73 watts.

  • 79 frames per second with VGG-19, a leading classification model (224x224x3) at 3.6 watts.

To learn more about the gluon, visit AlphaICs https://www.alphaics.ai or write to [email protected].

About AlphaICs:

AlphaICs is a leading AI technology company developing edge inference and edge learning technologies to enable AI at the edge. AlphaICs has a next-generation AI architecture called Real AI Processor (RAPTM). The architecture offers high performance, low power consumption and minimal latency, enabling best-in-class edge AI inference processors. RAPTM The architecture also supports edge learning to reduce training data requirements, enabling automatic labeling and continuous learning at the edge. The company is led by a team of technology experts and successful serial entrepreneurs dedicated to unlocking the true potential of AI at the edge. The company has offices in Milpitas, USA and Bangalore, India. AlphaICs is currently a company in the Silicon Catalyst Incubator. Learn more at https://www.alphaics.ai

Related Images

Figure 1: AlphaICs unveils the Gluon chip and Gluon board

AlphaICs’ Pradeep Vajram and Prashant Trivedi present Gluon chip and Gluon board

This content was published by the News release distribution service at Newswire.com.

Appendix

Share.

Comments are closed.