Intel Unveils New AI Chip to Compete with Nvidia and AMD

Last Updated on December 15, 2023 by SPN Editor

Intel launched its new AI Chip “AI Everywhere” initiative, revealing an unparalleled array of AI products designed to enable AI solutions across various platforms – from data centers and clouds to networks, edge devices, and PCs. In collaboration with partners and a vast ecosystem, Intel is paving the way for new growth opportunities powered by AI, embodying the concept of AI everywhere.

The Dawn of the New AI PC Chip with Intel Core Ultra

The Intel Core Ultra mobile processor family, a first of its kind built on Intel 4 process technology, marks the company’s most significant architectural shift in four decades. This new processor family, which is the most power-efficient client processor from Intel, heralds the advent of the AI PC era.

5th Gen Intel Xeon: AI Acceleration in Every Core

The 5th Gen Intel Xeon processor family, equipped with AI acceleration in every core, promises substantial improvements in AI and overall performance while reducing the total cost of ownership (TCO).

Intel Gaudi 3 AI Accelerator: Coming Soon

Intel CEO Pat Gelsinger unveiled the Intel Gaudi 3 AI accelerator, set to launch as scheduled next year.

Intel’s Vision for New AI Chip

“AI innovation is set to increase the digital economy’s contribution to as much as one-third of the global gross domestic product,” Gelsinger stated. He emphasized Intel’s commitment to developing technologies and solutions that enable customers to seamlessly integrate and effectively deploy new AI in all their applications, whether in the cloud or locally at the PC and edge where data is generated and used.

Gelsinger highlighted Intel’s extensive AI footprint, which spans cloud and enterprise servers, networks, volume clients, and ubiquitous edge environments. He also confirmed that Intel is on track to deliver five new process technology nodes in four years.

Intel’s Mission: AI Everywhere

“Intel is on a mission to bring the Intel launched its new AI Chip “AI everywhere” through exceptionally engineered platforms, secure solutions, and support for open ecosystems. Our AI portfolio gets even stronger with today’s launch of Intel Core Ultra, ushering in the age of the AI PC and AI-accelerated 5th Gen Xeon for the enterprise,” Gelsinger said.

Intel Core Ultra: Powering AI PCs and New Applications

Intel Core Ultra, representing the company’s largest architectural shift in 40 years, launches the AI PC generation with innovation on all fronts: CPU compute, graphics, power, battery life, and profound new AI features. The AI PC represents the largest transformation of the PC experience in 20 years, since Intel Centrino untethered laptops to connect to Wi-Fi from anywhere.

Intel Core Ultra features Intel’s first client on-chip AI accelerator — the neural processing unit, or NPU — to enable a new level of power-efficient AI acceleration with 2.5x better power efficiency than the previous generation. Its world-class GPU and leadership CPU are each also capable of speeding up AI solutions.

Intel is partnering with more than 100 software vendors to bring several hundred AI-boosted applications to the PC market — a wide array of highly creative, productive, and fun applications that will change the PC experience. For consumer and commercial customers, this means a larger and more extensive set of AI-enhanced applications will run great on Intel Core Ultra, particularly compared to competing platforms. For example, content creators working in Adobe Premiere Pro will enjoy 40% better performance versus the competition.

Intel Core Ultra-based AI PCs are available now from select U.S. retailers for the holiday season. Over the next year, Intel Core Ultra will bring AI to more than 230 designs from laptop and PC makers worldwide. AI PCs will comprise 80% of the PC market by 2028 and will bring new tools to the way we work, learn, and create.

The New Xeon: Amplifying AI Capabilities Across Various Platforms

Today, the 5th Gen Intel Xeon processor family was unveiled, marking a significant stride in performance and efficiency. When compared to its predecessor, these processors offer a 21% average performance boost for general compute performance and enable a 36% higher average performance per watt across a spectrum of customer workloads. Customers who adhere to a typical five-year refresh cycle and upgrade from even older generations can decrease their Total Cost of Ownership (TCO) by up to 77%.

The Xeon stands out as the only mainstream data center processor with built-in AI acceleration. The new 5th Gen Xeon delivers up to 42% higher inference and fine-tuning on models as large as 20 billion parameters. It’s also the only CPU with a consistent and ever-improving set of MLPerf training and inference benchmark results.

The built-in AI accelerators of Xeon, coupled with optimized software and enhanced telemetry capabilities, facilitate more manageable and efficient deployments of demanding network and edge workloads. This is beneficial for communication service providers, content delivery networks, and a wide range of vertical markets, including retail, healthcare, and manufacturing.

IBM announced that the 5th Gen Intel Xeon processors achieved up to 2.7x better query throughput on its WatsonX data platform compared to previous-generation Xeon processors during testing.

Google Cloud, which will deploy the 5th Gen Xeon next year, noted that Palo Alto Networks experienced a 2x performance boost in its threat detection deep learning models by using built-in acceleration in 4th Gen Xeon through Google Cloud. Indie game studio Gallium Studios leveraged Numenta’s AI platform running on Xeon processors to improve inference performance by 6.5x over a GPU-based cloud instance, reducing cost and latency in its AI-based game, Proxi.

This level of performance paves the way for advanced AI possibilities – not just in the data center and cloud, but across global networks and edge applications.

AI Acceleration: Meeting Developers’ Needs Everywhere

Both Intel Core Ultra and 5th Gen Xeon are set to infiltrate unexpected places. Envision a restaurant that tailors your menu choices based on your budget and dietary requirements; a manufacturing floor that detects quality and safety issues at the source; an ultrasound that perceives what human eyes might overlook; and a power grid that manages electricity with meticulous precision.

These edge computing use cases represent the fastest-growing segment of computing — projected to skyrocket to a $445 billion global market by the end of the decade — within which AI is the fastest-growing workload. In this market, edge, and client devices are driving 1.4x more demand for inference than the data center.

In many instances, customers will utilize a combination of AI solutions. For example, Zoom runs AI workloads on Intel Core-based client systems and Intel Xeon-based cloud solutions within its comprehensive communications and collaboration platform to deliver optimal user experience and costs. Zoom employs AI to suppress the sound of a neighbor’s barking dog, blur your cluttered home office, and generate a meeting summary and email.

To make AI hardware technologies as user-friendly and accessible as possible, Intel incorporates optimizations into the AI frameworks developers use (like PyTorch and TensorFlow) and provides foundational libraries (through oneAPI) to make software portable and highly performant across different types of hardware.

Advanced developer tools, including Intel’s oneAPI and OpenVINO toolkit, assist developers in harnessing hardware acceleration for AI workloads and solutions, and swiftly build, optimize, and deploy AI models across a broad range of inference targets.

Intel Gaudi3 AI Accelerator: A Glimpse into the Future

Gelsinger offered an insight into the future of AI with Intel Gaudi3, set to make its debut next year. For the first time, he unveiled the upcoming AI accelerator, designed specifically for deep learning and large-scale generative AI models.

Intel’s Gaudi pipeline has witnessed a swift expansion, thanks to its growing performance benefits coupled with a highly competitive Total Cost of Ownership (TCO) and pricing structure. As the demand for generative AI solutions continues to rise, Intel anticipates securing a larger share of the accelerator market in 2024, led by its suite of AI accelerators, with Gaudi at the forefront.