Missed a session at the Data Summit? View on demand here.

The market for edge AI chips, designed to accelerate AI workloads offline, often in standalone hardware, is maturing at a rapid pace. Building on converging trends such as digital transformation, cloud-native technologies and the Internet of Things, the edge AI hardware segment could be worth as much as $38.87 billion by 2030, according to an estimate by Valuates Reports.

Valuates cites increased demand for low latency and real-time processing, as well as reductions in storage and operational costs, as factors driving the adoption of edge AI chips. Indeed, these kinds of chips can enable better performance and lower power consumption by reducing the need for devices to rely on the cloud for data processing. But edge AI chips are chained in other ways: For example, because they don’t have the computing power of, say, a cloud data center, only certain tasks can be performed on an edge device.

Quadric is one of the many startups entering the AI ​​edge market with fervor, promising to eliminate the historical bottlenecks of edge hardware. Today, Quadric announced a $21 million Series B financing round co-led by Denso’s NSITEXE and MegaChips, with the participation of Leawood VC, Pear VC, Uncork Capital and Cota Capital, to bolster production of its edge AI chips, which the company claims it can. “accelerate the entire application pipeline” on the device without the need for a powerful general purpose processor.

Secret fries sauce

Based in Burlingame, California, Quadric was founded in 2016 by Veerbhan Kheterpal, Nigel Drego and Daniel Firu. All three come from MIT and Carnegie Mellon and previously co-founded cryptocurrency computing company 21 Inc.

“The founding team was building a smart robot when faced with the inadequacy of existing computing platforms from Nvidia and Intel,” a Quadric spokesperson told VentureBeat via email. “Unless rebuilt from the ground up, processors used for edge computing are not scalable. Quadric was founded to build a new processor architecture; one that generalizes the data flow paradigm and delivers a higher level of energy efficiency for a wide range of algorithms in machine learning, computer vision, DSP, graphics processing and linear algebra.”

Quadric claims that its 1.1 billion transistor, 16 nanometer chip consumes only 4.5 W of power and contains 4 GB of memory combined with 256 “vertex cores,” which are designed to handle some of the algorithmic workloads involved in general operations. Accelerate AI applications. The workloads don’t include training or the step of developing an AI system that needs to feed the system massive amounts of data so it learns to make predictions. Rather, it’s about inference, the point at which the system can make predictions based on new data coming in.

“Quadric’s unique ability to handle both neural backbones and classic dynamic data-parallel algorithms in a unified architecture helps create AI for everyone, everywhere. Most other solutions combine powerful processor clusters with application-specific neural processing units,” said the Quadric spokesperson. The company further explains on its website: “The architecture is instruction-driven… In conjunction with” [it] is a software programming model tailored for the ease of use of developers. The software programming model allows the developer to express graph-based and non-graph-based algorithms in harmony.

Quadric provides plug-and-play AI models for applications in the warehouse, construction, transportation and agriculture industries. The company previously claimed that Denso plans to integrate its edge chip technology, which works with any machine with an M.2 motherboard expansion slot, into future self-driving vehicle solutions.

Expanding edge market

Deloitte estimates that more than 750 million edge AI chips that perform tasks on the device have been sold to date, bringing in $2.6 billion in revenue.

“The volume of data generated in enterprises is growing rapidly, so to handle these volumes of data, the next generation of innovation in computing will take place outside the data center and closer to the network edge,” the spokesperson added. “Quadric helps enterprises create privacy-sensitive data solutions and optimize latency and bandwidth costs.”

Quadric competes with companies such as AI Storm, Axelera, Deep Vision, Flex Logix, Sima.ai, Blaize and Hailo, the latter of which has raised more than $320 million at a valuation reportedly exceeding $1 billion. As ZDNet’s Tiernan Ray highlighted in a recent piece, venture financing has boosted the AI ​​edge chip market, with dozens of vendors (one report counted more than 60) vying for a slice of the growing cash pile.

But Quadric believes its AI chip architecture sets it apart in the fast-growing space. The company insists its investment will enable Quadric, which reportedly has five customers, to release the next version of its chip architecture; improve the performance of the software development kit that comes with the chips; and roll out new products for integration into system-on-chips.

“Most other companies in the edge computing space are building dedicated workload accelerators. Quadric’s software-centric architecture, on the other hand, is future-proof against a dynamic backdrop of algorithms and AI models,” explains the spokesperson.

To date, Quadric has raised $34 million in venture capital and $2 million in debt with 35 employees, including a round of $15 million in May 2019.

VentureBeat’s mission is to be a digital city square for tech decision makers to learn about transformative business technology and transactions. Learn more

This post Quadric raises $21 million to accelerate production of its AI edge chips was original published at “https://venturebeat.com/2022/03/16/quadric-nabs-21m-to-accelerate-production-of-its-ai-edge-chips/”