Untether AI Selected by Ola-Krutrim for Next-Gen AI Acceleration Solutions

TORONTO, Aug. 16, 2024 — Untether AI, a leader in energy-centric AI inference acceleration, has announced its partnership with Ola-Krutrim to provide AI acceleration solutions for its India data centers. This multi-faceted partnership includes today’s performance demonstration with speedAI-based acceleration solutions and co-development of Untether AI’s next generation datacenter solutions, which will be used for inference deployment and fine-tuning of Krutrim’s generative AI models. Untether AI was selected by Krutrim because its energy-efficient At-Memory compute is ideally suited for the high-performance, high-scalability required for the Indian market.

Today at Ola Electric’s Sankalp 2024 event the two companies demonstrated an implementation using an Arm-based host CPU and the single-chip Untether AI speedAI acceleration card, providing higher throughput, better energy efficiency, and significantly lower cost compared to current GPU-based solutions.

“Today’s announcement and demonstration begins our partnership with Ola-Krutrim to build upon the world-class performance and energy efficiency in AI inference that Untether AI provides,” said Chris Walker, CEO of Untether AI. “We are gratified to be working in close collaboration with Ola-Krutrim’s hardware and software engineering teams to develop next generation AI acceleration solutions utilizing the capability of our At-Memory speedAI architecture.”

“Amidst rapidly changing technology paradigms, India has a unique opportunity to become a global superpower in technologies of the future such as AI. India being the source for 20% of the world’s data and home to one of the largest tech talent pool, is fully equipped to lead the AI wave,” said Bhavish Aggarwal, Founder and CEO of Ola Krutrim. “With Krutrim, we are focused on making India technologically independent and are making rapid progress in developing full-stack AI capabilities, silicon upwards.”

With this announcement, Untether AI extends the speedAI architecture from the edge to the cloud enabling scalable AI performance. This also expands Untether AI’s capabilities in delivering innovative solutions in collaboration with other industry leaders.

About Untether AI

Untether AI provides energy-centric AI inference acceleration from the edge to the cloud, supporting any type of neural network model. With its at-memory compute architecture, Untether AI has solved the data movement bottleneck that costs energy and performance in traditional CPUs and GPUs, resulting in high-performance, low-latency neural network inference acceleration Untether AI delivers its technology in speedAI devices and acceleration cards, and its imAIgine Software Development Kit. More information can be found at www.untether.ai.

Source: Untether AI