Gyrfalcon's production-ready AI accelerator with embedded MRAM
Gyrfalcon Technology Inc. (GTI) announced a production-ready 22 nm application-specific integrated circuit (ASIC) chip with embedded magnetoresistive random-access memory (MRAM), the Lightspeeur 2802M. It was delivered to GTI by TSMC in June of 2018, and the company has been preparing to discuss the new offering with customers interested in designing solutions for IoT endpoints, cloud solutions, autonomous vehicles, and other use cases.
The company says that its 2802M is the first AI accelerator to deliver the benefits of MRAM, such as non-volatility and low power, as well as significant advancements specific to Edge AI. The 2802M includes 40 MB of memory that can support large AI models or multiple AI models within a single chip. Examples of models that could be supported within the same chip are image classification, voice identification, voice commands, facial recognition, and pattern recognition.
GTI says it is looking forward to working on large-scale opportunities with customers to develop specific "super AI accelerator chips" that combine the new MRAM engine with its other patented innovations such as the MPE and APiM technologies. The MPE (matrix processing engine) uses CNN with the APiM (AI processing in memory), pairing logic with memory like a real human brain. This optimizes the speed of processing, achieving high TOPS performance, while also saving tremendous amounts of power by avoiding management of data in discrete memory components.
GTI will present its technologies, solutions, and chips at CES in January 2019. For more information and to plan your visit, go to https://www.gyrfalcontech.ai/ces-2019/.