Untether AI Partners with Colfax Worldwide to Offer Peak Functionality in AI Edge Servers

TORONTO–(Company WIRE)–Feb 4, 2021– Untether AI, the chief in at-memory computation and AI acceleration, currently…

TORONTO–(Company WIRE)–Feb 4, 2021–

Untether AI, the chief in at-memory computation and AI acceleration, currently announced a partnership with Colfax Intercontinental in which Untether’s tsunAImi TM accelerator cards, run by the runAI TM units, are now offered in the Colfax AIEdge™ servers. Untether AI’s tsunAImi accelerator playing cards enable corporations jogging AI workloads to operate neural networks faster, cooler, and a lot more cost-successfully working with at-memory computing. The cards, alongside with Untether AI’s imAIgine Software program Advancement Kit (SDK), gives AI developers an automatic route to managing networks at significant effectiveness. This frees them from owning to perform reduced-degree optimization tasks and instead spend time the place it issues to them – crafting their styles.

“Untether AI’s ground breaking method, using at-memory computation, will help Colfax to offer unmatched compute density to support buyers realize the most affordable TCO for inference workloads. We are fired up to commence our journey with Untether AI,” said Gautam Shah, CEO of Colfax.

Artificial Intelligence (AI) workloads for inference involve expanding amounts of compute sources, much outstripping the gains offered to standard CPU and GPU architectures. Untether AI radically rethought how computation for device discovering is attained, by concentrating on the desires for inference acceleration and maximizing power performance. Using at-memory computation, Untether AI breaks as a result of the obstacles of regular von Neumann architectures, supplying sector-primary, cost-successful compute density with power efficiency.

Colfax presents highly developed computing alternatives for those at the forefront of innovation and discovery and partners with some of the major manufacturer names in the business. The blend of Untether AI’s tsunAImi acceleration card and Colfax servers gives the industry’s optimum compute density for AI inference workloads.

“Over the training course of several a long time, Colfax has attained a reputation for resolving the most sophisticated troubles and securing exceptional results for enterprises, countrywide labs, universities, general public businesses and get started-ups. We are happy to have Colfax as a lover,” stated Bob Beachler, VP of Item, Untether AI.


Untether AI tsunAImi card

With the industry’s greatest compute density and very best performance-per-watt, Untether AI tsunAImi playing cards permit Colfax to custom-configure servers for inference workloads that can cut down their money and operational expenses by up to 75%. tsunAImi accelerator cards are powered by 4 runAI200 devices, offering 2 PetaOperations of compute, more than two times other out there card. This compute energy interprets into more than 80,000 frames for every 2nd of ResNet-50 v 1.5 throughput at batch=1, three instances the throughput of its nearest competitor. For all-natural language processing, tsunAImi accelerator cards can process a lot more than 12,000 queries per second (qps) of BERT-foundation.

The imAIgine Computer software Enhancement Package

The Untether AI imAIgine Software program Improvement Package (SDK) delivers an automatic route to running higher-functionality networks, thanks to drive-button quantization, optimization, physical allocation, and multi-chip partitioning. The imAIgine SDK also presents an substantial visualization toolkit, cycle-accurate simulator, and an quickly integrated runtime API.

For more info on the Colfax AI Edge Servers and availability visit Colfax International.

About Untether AI

Untether AI supplies extremely-economical, high-general performance AI chips to enable new frontiers in AI purposes. By combining the power performance of at-memory computation with the robustness of electronic processing, Untether AI has developed a groundbreaking new chip architecture for neural web inference that removes the facts motion bottleneck that expenses power and effectiveness in conventional architectures. Started in Toronto in 2018, Untether AI is funded by Radical Ventures and Intel Capital. www.untether.ai.

All reference to Untether AI trademarks are the assets of Untether.AI.
All other logos pointed out herein are the home of their respective house owners.

See resource variation on businesswire.com:https://www.businesswire.com/news/dwelling/20210204005099/en/

Make contact with: Media Get hold of for Untether AI:

Michelle Clancy, Cayenne Worldwide, +1.503.702.4732

[email protected] Contact:

Robert Beachler, Untether AI, +1.650.793.8219

[email protected] Speak to:

Manish [email protected] with Untether AI:

Twitter: @UntetherAI

LinkedIn: https://www.linkedin.com/company/untether-ai/

Keyword: NORTH The us CANADA

Sector Key phrase: TELECOMMUNICATIONS Software program NETWORKS World-wide-web Components Digital Layout AUTOMATION Knowledge Management Technologies

Resource: Untether AI

Copyright Small business Wire 2021.

PUB: 02/04/2021 11:00 AM/DISC: 02/04/2021 11:01 AM

http://www.businesswire.com/news/property/20210204005099/en