• Skip to main content
  • Skip to secondary menu
  • Skip to footer

Technologies.org

Technology Trends: Follow the Money

  • Technology Events 2026-2027
  • Sponsored Post
  • Technology Markets
  • About
    • GDPR
  • Contact

Intel Unveils New GPUs and AI Accelerators at Computex 2025

May 19, 2025 By admin Leave a Comment

Celebrating 40 Years of Innovation in Taiwan

At Computex 2025, Intel made significant announcements highlighting its ongoing commitment to professional developers and the expanding artificial intelligence ecosystem. Marking its 40th anniversary of operations in Taiwan, Intel introduced several new products that demonstrate continued evolution in GPU technology and AI acceleration capabilities, emphasizing its deep-rooted collaboration with Taiwan’s vibrant technology landscape. Among the unveiled products are the Intel Arc Pro B-Series GPUs, Intel Gaudi 3 AI accelerators, and the public beta release of the Intel AI Assistant Builder, designed to empower creators, developers, and enterprises with advanced, scalable, and efficient tools.

The newly announced Intel Arc Pro B60 and B50 GPUs represent significant additions to Intel’s professional graphics lineup, built on the Xe2 architecture optimized for high-performance AI inference tasks and professional workstation applications. Equipped with Intel Xe Matrix Extensions (XMX) AI cores and state-of-the-art ray tracing units, these GPUs cater specifically to professionals working in architecture, engineering, construction (AEC), and AI development sectors. The Arc Pro B60 GPU boasts 24GB of memory, while the Arc Pro B50 GPU provides 16GB, making them suitable for intensive workloads demanding considerable memory resources. Intel also showcased Project Battlematrix, a workstation-class Xeon-based platform capable of supporting up to eight Arc Pro B60 GPUs, offering a robust environment capable of training and running medium-sized, highly accurate AI models with up to 192GB of video RAM.

Recognizing the practical needs of professional users, Intel has ensured broad compatibility across software ecosystems. The GPUs feature certifications from various independent software vendors (ISVs), ensuring stability and optimized performance across professional applications. On Windows, they offer dual compatibility with consumer and professional drivers, while Linux users benefit from a containerized software stack designed for simplified AI deployment and continued optimization. Intel aims to expand availability swiftly, with samples of the Arc Pro B60 GPU set to ship from board partners like ASRock, Gunnir, and Sparkle beginning June 2025, while the Arc Pro B50 GPU will reach resellers by July 2025.

Complementing the GPU announcements, Intel revealed the expanded capabilities of its Gaudi 3 AI accelerators, now available in both PCIe and rack scale formats. The Gaudi 3 PCIe cards offer significant flexibility and scalability, ideal for deployment within existing data center server infrastructures. With support for a wide range of AI models, from small-scale models such as Llama 3.1 8B to larger, advanced models like the full-scale Llama 4 Scout or Maverick, these accelerators aim to empower enterprises of all sizes to meet their evolving AI inference demands effectively. The Gaudi 3 rack scale system reference designs further amplify Intel’s commitment to providing open, scalable AI infrastructure. Capable of supporting up to 64 accelerators per rack, these modular designs integrate high-bandwidth memory totaling 8.2TB and liquid cooling systems to maximize performance and maintain operational cost efficiency.

Additionally, Intel made the Intel AI Assistant Builder publicly accessible on GitHub, enabling developers and businesses to create custom AI agents optimized for Intel-powered devices. First introduced at CES 2025, this software framework simplifies the process of developing localized AI solutions tailored to specific organizational or customer needs. Demonstrations from Intel’s partners Acer and ASUS at Computex illustrate practical applications and the platform’s versatility in streamlining AI assistant development and deployment.

These announcements not only showcase Intel’s latest innovations but also highlight the importance of Intel’s longstanding relationship with Taiwan’s tech ecosystem, emphasizing how collaborative innovation continues to propel technology forward. Intel CEO Lip-Bu Tan underscored this relationship, acknowledging four decades of partnership that have consistently delivered impactful advancements in computing technology. As Intel moves forward, its newly launched products signify a renewed pledge to drive technological progress in AI, GPU computing, and collaborative innovation, shaping the future of professional computing and enterprise-grade AI solutions globally.

Filed Under: News

Reader Interactions

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Footer

Recent Posts

  • Why ServiceNow, Salesforce, and Atlassian Are Selling Off—and Whether the AI Fear Is Overdone
  • Infleqtion Rings the NYSE Bell
  • Temporal Raises $300M Series D at $5B Valuation to Take Agentic AI Into Production
  • ChipAgents Secures $50M Series A1, Reinforcing the Shift Toward Agentic AI in Chip Design
  • Render Raises $100M Series C Extension, Valued at $1.5B, Betting Big on the AI-Native Cloud
  • Uptiq Raises $25M Series B to Push Financial AI Out of the Demo Trap
  • From Desk to Flight: High-Value 3D Printing Ideas for a Home Premise
  • Positron AI Raises $230M Series B, Redefines the Economics of AI Inference
  • What You Can Build in Loveable, and Why It Feels Different
  • Forrester Sees Global Tech Spending Hitting $5.6 Trillion in 2026 as AI Drives Growth Despite Tariffs

Media Partners

  • Market Analysis
  • Cybersecurity Market
Accrual Launches With $75M to Push AI-Native Automation Into Core Accounting Workflows
Europe’s Digital Sovereignty Moment, or How Regulation Became a Competitive Handicap
Palantir Q4 2025: From Earnings Beat to Model Re-Rating
Baseten Raises $300M to Dominate the Inference Layer of AI, Valued at $5B
Nvidia’s China Problem Is Self-Inflicted, and Washington Should Stop Pretending Otherwise
USPS and the Theater of Control: How Government Freezes Failure in Place
Skild AI Funding Round Signals a Shift Toward Platform Economics in Robotics
Saks Sucks: Luxury Retail’s Debt-Fueled Mirage Collapses
Alpaca’s $1.15B Valuation Signals a Maturity Moment for Global Brokerage Infrastructure
The Immersive Experience in the Museum World
Booz Allen Hamilton Acquires Defy Security to Scale Commercial Cybersecurity
Cloudflare snd Mastercard: Closing the Cyber Resilience Gap for the Internet’s Most Exposed Organizations
CyberBay Summit 2026, March 12–13, Tampa Bay
VulnCheck Raises $25M Series B to Accelerate Machine-Speed Exploit Intelligence
CyberCube Appoints Chris Methven as CEO, Signaling Next Phase of Growth
Modveon Raises $10M to Build a Verified Operating System for Governments and Citizens
Modirum Platforms Joins Digital Defence Ecosystem Finland to Expand Europe’s Secure Digital Defence Capabilities
Salt Typhoon Reaches Scandinavia: When Telecom Espionage Goes Public in Norway
SentinelOne Expands AI Security to the First Mile, Redefining How Enterprises Protect AI Systems
NETSCOUT SYSTEMS Q3 FY2026: Quiet Acceleration, Better Mix, and a Cautious Turn Toward Growth

Media Partners

  • Market Research Media
  • Technology Conferences
When the Market Wants a Story, Not Numbers: Rethinking AMD’s Q4 Selloff
BBC and the Gaza War: How Disproportionate Attention Reshapes Reality
Parallel Museums: Why the Future of Art Might Be Copies, Not Originals
ClickHouse Series D, The $400M Bet That Data Infrastructure, Not Models, Will Decide the AI Era
AI Productivity Paradox: When Speed Eats Its Own Gain
Voice AI as Infrastructure: How Deepgram Signals a New Media Market Segment
Spangle AI and the Agentic Commerce Stack: When Discovery and Conversion Converge Into One Layer
PlayStation and the Quiet Power Center of a $200 Billion Gaming Industry
Adobe FY2025: AI Pulls the Levers, Cash Flow Leads the Story
Canva’s 2026 Creative Shift and the Rise of Imperfect-by-Design
Chiplet Summit 2026, February 17–19, Santa Clara Convention Center, Santa Clara, California
MIT Sloan CIO Symposium Innovation Showcase 2026, May 19, 2026, Cambridge, Massachusetts
Humanoid Robot Forum 2026, June 22–25, Chicago
Supercomputing Asia 2026, January 26–29, Osaka International Convention Center, Japan
Chiplet Summit 2026, February 17–19, Santa Clara Convention Center, Santa Clara, California
HumanX, 22–24 September 2026, Amsterdam
CES 2026, January 7–10, Las Vegas
Humanoids Summit Tokyo 2026, May 28–29, 2026, Takanawa Convention Center
Japan Pavilion at CES 2026, January 6–9, Las Vegas
KubeCon + CloudNativeCon Europe 2026, 23–26 March, Amsterdam

Copyright © 2022 Technologies.org

Media Partners: Market Analysis & Market Research and Exclusive Domains, Photography