Groq to Showcase World's Fastest Large Language Model Performance, Powered by Its LPU™ System, at the Global Emerging Technology Summit in Washington, DC

News provided by

Sep 20, 2023, 08:00 ET

The Groq LPU™ system is purpose-built and software-driven to power Large Language Models for the exploding AI market.
Groq showcases Llama-2 70B performance generating text faster than 240 tokens per second per user—the world's fastest generation speed of a foundational Large Language Model.
Groq CEO Jonathan Ross will demonstrate the LPU™ system in his talk, "Unleashing the Power of Large Language Models," at the Global Emerging Technology Summit in Washington, DC, to be livestreamed on September 21^st at 10:25am EDT

MOUNTAIN VIEW, Calif., Sept. 20, 2023 /PRNewswire/ -- Groq, an AI solutions company announced today a record-breaking AI processing demo, powered by the ultra-low latency performance of their LPU™ system, to be delivered by CEO and Founder Jonathan Ross tomorrow, September 21st, at the 2023 Global Emerging Technology Summit, hosted by the Special Competitive Studies Project in Washington DC. The Groq LPU™ system running Llama-2 70B produces generated text at a speed of more than 240 tokens per second per user—the world's fastest performance on any foundational Large Language Model (LLM) of this scale.

Groq to showcase the world’s fastest large language model performance, powered by its LPU™ system, at the Global Emerging Technology Summit in Washington, DC.

Jonathan Ross, CEO and Founder of Groq, presenting the latest Llama-2 70B Large Language Model record breaking performance to an audience in Mountain View, CA at GroqDay on September 7, 2023.

Performance, efficiency, speed, and accuracy are packed into GroqChip™–the basis of the Groq Language Processing Unit™ system–which is simpler in design and layout than a Graphics Processing Unit (GPU), faster than a GPU, and less expensive than a GPU. AI powered by Groq will enable a "language user interface" that allows anyone to create AI applications faster than ever before.

"The demands for AI are accelerating and GPUs can't handle the speed and the accuracy required," said Jonathan Ross, Groq CEO and Founder. "The GPU is becoming the weakest link in the AI economy. We intend to set a new standard for what the AI experience should be, while making it accessible to everyone."

Ross will be sharing the SCSP stage with other luminaries in the AI field, helping to support the US in its efforts to harness generative AI for global competitiveness. Register here to attend the event virtually. Schedule an exclusive technology demo at [email protected].

About Groq
Groq is an AI solutions company and the inventor of the Language Processing Unit™ accelerator that is purpose-built and software-driven to power Large Language Models (LLMs) for the exploding AI market. For more information, visit www.groq.com.

Groq, the Groq logo, and other Groq marks are trademarks of Groq, Inc. Other names and brands may be claimed as the property of others. Reference to specific trade names, trademarks or otherwise, does not necessarily constitute or imply its endorsement or recommendation by Groq.

About SPSC
SCSP is a non-partisan, non-profit initiative whose mission is to make recommendations to strengthen America's long-term competitiveness as AI reshapes US national security, economy, and society.

CONTACT: [email protected]

SOURCE Groq