Public Sector Hero Image

Bringing Speed to Mission with the
Groq LPU™ Inference Engine

Groq builds the world’s fastest AI inference technology. The LPU™ Inference Engine by Groq is a hardware and software platform that delivers exceptional compute speed, quality, and energy efficiency.

Groq, headquartered in Silicon Valley, provides cloud and on-prem solutions at scale for AI applications. The LPU and related systems are designed, fabricated, and assembled in North America.

Our primary benefits are superior performance, pace, and energy efficiency.

Groq solutions provide faster time-to-market for deploying inference workloads with far less complexity and cost. Our kernel-less compiler can process most workloads in a small fraction of the time of GPU-based inference systems–days, not months–and requires far fewer engineers. This not only accelerates the pace of solution development and deployment, it solves the human capital problem. With Groq, you need fewer people to deploy and scale workloads.

Groq solutions, including our software tools and deterministic architecture, enable developers to speed up their production deployment and reduce time to insights for mission critical objectives. Developers are provided with key metrics for production deployments at compile time, ensuring predictable and repeatable performance metrics at scale. With Groq, human capital can take the time they’re given back to focus on innovating and advancing, rather than deploying.

For large scale inference, Groq is simply faster, providing an optimized end-to-end system needed to glean real-time insights ensuring the intelligence analyst or the warfighter gets what they need, when they need it. How much faster depends on a number of factors, but in many cases it’s more than 10X, giving the US safety and security advantages over adversarial nations.

Our TruePoint™ technology maintains accuracy while exploiting the efficiency of lower precision, further enhancing performance.

Groq can help agencies address their enduring need for higher performance, lower latency compute solutions to process large volumes of data faster, while using less power. The LPU Inference Engine runs GenAI applications at 10X better speed and precision, opening the door to an entirely new class of real-time AI solutions. The Public Sector needs our unique approach to inference to meet mission critical objectives when time and accuracy counts most for our citizens.

The official partnership between Groq and Carahsoft Technology Corp., The Trusted Government IT Solutions Provider®, will deliver the fastest and cost- and energy-efficient AI inference speed to Government agencies and Federal systems integrators throughout the United States. Under the distribution agreement, Carahsoft will serve as Public Sector distributor for Groq, making its innovative AI inference solutions available to the Public Sector through Carahsoft’s reseller partners and NASA Solutions for Enterprise-Wide Procurement (SEWP) V contracts. GroqCloud and Groq private cloud offerings are available through Carahsoft’s SEWP V Contracts NNG15SC03B and NNG15SC27B, and ITES-SW2 Contract W52P1J-20-D-0042.

News & Resources


Contact Us

Jonathan Ross

CEO, Founder

Jonathan Ross is the CEO and founder of Groq Inc. Groq is the creator of the world’s first Language Processing Unit (LPU), providing exceptional speed performance for AI workloads running on their LPU™ Inference Engine.

Prior to founding Groq, Jonathan began what became Google’s Tensor Processing Unit (TPU) as a 20% project where he designed and implemented the core elements of the first-generation TPU chip. Jonathan next joined Google X’s Rapid Eval Team, the initial stage of the famed “Moonshots factory”, where he devised and incubated new Bets (Units) for Google’s parent company, Alphabet.

Adam Tachner

CLO & VP of Corporate Development

For nearly 20 years, Adam Tachner has provided high-growth semiconductor companies with the cross-functional business leadership they need to thrive in competitive markets. As head of the legal and compliance functions at Atheros Communications (acquired 2011 by Qualcomm for $3.6B), Adam supported the company’s rapid growth from a pre-revenue innovator to a market-leading IPO and a $1B annual revenue run rate. Later, he helped integrate Atheros into Qualcomm’s global operations. Most recently, Adam held both VP Corporate Development and General Counsel roles at Google’s Access division, and earlier at fabless MEMS leader InvenSense, now the sensor division of TDK.
Estelle Hong -- Chief of Staff to CEO & Head of Strategic Operations copy
Estelle Hong

Head, Planning & Execution (PLEX)

Estelle is a jack-of-all-trades by day, and a National Guardsman by night, bringing more than a decade of leadership experience from the Army and Intel Corporation. Estelle graduated from West Point with an Engineering Management degree and was commissioned as an Engineer Officer. Later, she received her MBA at Cornell University and joined Intel’s Accelerated Leadership Program, driving results in business HR, retail analytics, and the memory business. Estelle’s passion for driving disruptive technology and working with incredibly smart people aligns nicely in her opportunity to help scale Groq become an iconic company. Outside of work, she likes to snowboard, travel, play guitar, or hang out with her friends in San Francisco.
Aileen Black

President, Government & Public Sector

Aileen Black is a highly respected IT industry executive and general manager with a dynamic record of driving exponential growth results in public sector organizations. Her unique pedigree includes developing industry-changing technology and service businesses from the ground up, with proven experience in running large enterprise-class business units in excess of 600 million. Black has held executive positions at Google, VMware, EMC, and Oracle. With more than 30 years of experience building, managing, and motivating top-notch technology sales and professional services organizations, she has proven success with a deep understanding of the cloud, artificial intelligence, enterprise open source, big data, government contracting, sales, strategic alliances, marketing and the political landscape across the public sector market in addition to extensive media and public speaking across all forms of media including radio and television.
John Barrus
John Barrus

VP of Product

Barrus recently rejoined Groq after launching the Amazon Cardinal warehouse robot. An accomplished, high-tech leader, Barrus uses his deep technical background to drive the creation of new commercial products, including chips and handheld devices, room-size robotic workcells, and cloud accelerator infrastructure. Prior to joining Groq, Barrus led the launch of Google Cloud GPUs and TPUs, products that contribute significantly to Google’s $9B IaaS revenue. Educated at M.I.T., Barrus was the Director of the Ricoh Innovations Silicon Valley lab, has published 25 technical papers and journal articles, and has more than 80 issued U.S. patents. He is the founder of several startups and new business divisions.
Sunny Madra
Sunny Madra

GM, GroqCloud™

Sunny Madra, an experienced entrepreneur, has a track record of successful exits for the startups he founded. These exits include the sales of Autonomic to Ford Xtreme Labs to Pivotal, and most recently Definitive Intelligence to Groq. Currently, Sunny is leading GroqCloud.