Leader in Real-time AI Inference on Track to Deliver 50% of the World’s Inference Compute Capacity via GroqCloud™ by End of 2025
Groq, the only provider of real-time AI inference solutions for generative AI (GenAI), has signed a Letter of Intent with Earth Wind & Power to develop the first European vertically-integrated AI Compute Center, located in Norway. The second deal of its kind from Groq in the last two months, the deployment of the Groq LPU™ Inference Engine in Norway will provide Norwegians, European and NATO-allied countries with the lowest cost, most energy-efficient, scalable access to inference compute – the specific type of compute needed as the world shifts from training to powering GenAI applications. This AI Compute Center will also ignite the generative age economy in Norway and the EU by powering GenAI solutions and providing investors with an early opportunity to participate in the inference market.
In accordance with the terms of the Letter of Intent, Groq has committed to deploy and operate 21,600 LPUs at Earth Wind & Power’s AI Compute Center in 2024, with the option to increase this number to 129,600 LPUs in 2025.
“Besides the size, what makes this deployment particularly exciting is that it will require a fraction of the power to run vs. a GPU cluster. GPUs are notoriously thirsty for power, often requiring as much power as the average household per chip. LPUs use as little as a tenth as much power,” said Groq CEO and Founder Jonathan Ross. “The AI Compute Center in Norway will provide enough generative AI inference capacity to satisfy most of the needs of the country, and also offer European developers access to GDPR compliant inference infrastructure at the same scale that historically has only been available to US tech companies. This is just the start though, by the end of 2025 we intend to deploy more than 1.5 million LPUs world-wide, helping to democratize access to generative AI.”
Also Read: Windstream Wholesale Enhances Network Connectivity with Two New Data Center Expansions
For comparison, tech giants Meta and Microsoft are estimated to have deployed around 150,000 Nvidia H100 GPUs each, world-wide.
“It has become increasingly clear that the future is AI, which is why we are moving quickly to make inference compute accessible to developers, enterprises, and government entities who are at the cutting edge of GenAI in Norway and across Europe,” said Earth Wind & Power Co-Founder Ingvil Smines Tybring-Gjedde. “However, making AI progress can only happen while protecting energy resources and costs for Norwegian and EU citizens. Groq is the ideal partner to us in this as its LPU™ Inference Engine is the only solution available built to do just that. Groq and Earth Wind & Power also share the ethos that AI can and will play a significant role in ensuring free and fair elections around the globe and supporting the Ukrainian people in their fight for freedom. We look forward to working closely with Norwegian and NATO experts who are leading these efforts using GenAI today, as well as realizing the expected economic benefits for Norway.”
The signing took place Tuesday, April 16 at Groq office in Palo Alto, home to the GroqCloud team, with Earth Wind & Power and Norwegian business delegations in attendance. The day prior, H.R.H. Crown Prince Haakon of Norway and Jonathan Ross took the stage during the Business Norway event at the San Francisco Ferry Building which discussed the future of technology and AI. Ross was joined by Venture Capitalist Bill Tai and the Norway Minister of Digitalization Karianne Oldernes Tung for a conversation moderated by Tybring-Gjedde about the impacts of AI and the big-picture changes it may bring with an emphasis on business perspectives. Tai closed the session by noting the world “won’t have enough compute unless technologies like [Groq] come into play.”
Source: PRNewsWire