MOUNTAIN VIEW, Calif., Sept. 29, 2020 /PRNewswire/ -- Groq, Inc., the inventor of the unique Tensor Streaming Processor (TSP™) chip, announces that it has started shipping its latest Groq card, Groq node and Groq ware SDK solutions to customers worldwide.
The Groq node is a turnkey solution featuring eight Groq cards in a single chassis. Multiple Groq nodes can be connected in a wide variety of network topologies, enabling computation at a new scale. Each Groq node delivers up to 6 PetaOPs of performance, while consuming a mere 3.3KW of power in a 5U form factor. The high performance and low power of the Groq node offers unprecedented TCO (Total Cost of Ownership) benefits. Feedback confirms that Groq's solution delivers the shortest path to performance™.
Groq continues to ship its PCIe card featuring its TSP chip, adding industry standard server management capabilities. Groq's unique TSP is capable of 18,900 IPS (inferences per second) on ResNet-50 v2 at batch size one. This makes it the fastest commercially available AI/ML accelerator, with a responsiveness measured in hundredths of a millisecond. This eliminates the typical tradeoff between performance and latency. All this is testimony to the unique Groq architecture.
The announcement of Groq's hardware also includes the latest release of the Groq ware SDK, allowing developers to build high-performance models on Groq cards and nodes.
Andrew Fursman, Co-Founder and CEO at 1QBit said, "1QBit is pushing further into AI, and the bottlenecks we are running into are almost all of a computational nature. That's where Groq's solutions and architecture have helped us tremendously. We look forward to working further with Groq and are excited to see them expand their customer base in the healthcare, life sciences, and financial markets."
"Marvell is thrilled to enable the industry's first peta OP/s AI accelerator," said Igor Arsovski, CTO of the ASIC Business Unit at Marvell. "Marvell's unparalleled IP portfolio, including 90% higher-bandwidth dense SRAM, and our turnkey design center capability, enables our customers to focus on their own differentiated logic to deliver a product in record time. Our full architecture-technology co-optimization collaboration with Groq delivered a full-reticle AI accelerator with industry-leading performance from final netlist-to-tape-out in less than three months."
Groq is working with a number of global customers in the autonomous vehicle and financial spaces as well as a number of national labs, who are using the technology to address a wide variety of large compute challenges.
Groq Co-Founder & CEO, Jonathan Ross is presenting today at the virtual AI Hardware Summit, where he will provide additional information about the Groq card, node and Groq ware products.
Headquartered in Mountain View, CA, Groq delivers industry leading performance, accuracy and sub-millisecond latency with efficient, software-driven solutions for compute-intensive applications. Groq redefines compute by focusing on key technology innovations: software-defined compute, silicon innovation and developer velocity. For more information, visit: https://groq.com/
© Copyright 2020 Groq, Inc. Groq [if used, the Groq logo,] and TSP are trademarks of Groq. All other trademarks belong to their respective owners.