(Credit: Intel Corporation) New Intel Arc Pro B-Series GPUs: Intel announces Intel® Arc™ Pro B60 and Intel® Arc™ Pro B50 GPUs, expanding the Intel® Arc™ Pro family with configurations tailored for AI inference and professional workstations.
Intel AI Assistant Builder: Now publicly available on GitHub, Intel® AI Assistant Builder enables developers to create local, purpose-built AI agents optimized for Intel platforms.
Expanding Intel’s GPU lineup for professionals, new Intel Arc Pro B60 and Intel Arc Pro B50 GPUs are designed for demanding AI inference workloads and workstation applications.
Intel Gaudi 3 Supports PCIe and Rack Scale Deployments Expanding on its AI strategy, Intel introduced new deployment options for Intel Gaudi 3 AI accelerators: Intel Gaudi 3 PCIe cards support scalable AI inferencing within existing data center server environments.
Intel Booth at Computex 2025 Intel will showcase its latest processor and graphics innovations at Computex 2025 in the Taipei Nangang Exhibition Center.
The company announces the availability of the Intel Gaudi 3 AI accelerator via rack scale and PCIe deployments, and gives prosumers and AI developers access to the Intel Arc Pro GPU lineup.
TAIPEI, Taiwan–(BUSINESS WIRE)–Intel today introduces a new range of graphics processing units (GPUs) and AI accelerators for developers and professionals at Computex 2025. Among the announcements from Intel are…
Multimedia is used in this press release. You can view the complete release at https://www. businesswire . com/news/home/20250519565090/en/.
Intel expands the Arc Pro family with larger memory configurations and expanded software support with the release of the Arc Pro B60 and B50 graphics processing units for workstations and AI inference on May 19, 2025, ahead of Computex. At the May 20–23 event in Taipei, Taiwan, Intel celebrates 40 years of working with regional ecosystem partners and increases the availability of its GPU lineup, AI accelerators, and AI assistants. Intel Corporation deserves credit.
New Intel Arc Pro B-Series GPUs: Intel has announced the Intel® Arc™ Pro B60 and Intel® Arc™ Pro B50 GPUs, adding configurations specifically designed for professional workstations and AI inference to the Intel® Arc™ Pro family.
Currently available in PCIe and rack scale systems, Intel Gaudi 3 AI accelerators provide open, scalable solutions for cloud and enterprise AI inferencing.
With the Intel AI Assistant Builder, developers can now create locally tailored AI agents that are optimized for Intel platforms. It is now openly accessible on GitHub.
Taiwan’s 40th Anniversary.
The product launches this week mark Intel’s 40th anniversary in Taiwan, which marks 40 years of cooperation, innovation, and mutual success with one of the most vibrant technology ecosystems in the world centered around the x86 architecture. The x86 architecture continues to expand on its crucial role in boosting user and customer innovation as high-performance AI applications expand the demand for more processing power.
Additional: Intel Press Kit at Computex 2025.
“Over the last four decades, the strength of our collaboration with the Taiwan ecosystem has stimulated innovation that has improved our world,” stated Intel CEO Lip-Bu Tan. Our dedication to our partners is being reaffirmed this week as we strive to create a new Intel for the future. We will work together to develop fantastic products that satisfy our clients and take advantage of the exciting opportunities that lie ahead. “.”.
AI and workstation performance with new GPUs.
With their advanced ray tracing units and Intel® Xe Matrix Extensions (XMX) AI cores, the new Intel Arc Pro B60 and B50 GPUs, which are based on the Xe2 architecture, give engineers, developers, and creators high-performance capabilities.
New Intel Arc Pro B60 and Intel Arc Pro B50 GPUs are part of Intel’s professional GPU lineup and are made for workstation applications and demanding AI inference workloads. For creators, AI developers, and professionals, the Intel Arc Pro B-Series provides a flexible solution with AI-ready features, multi-GPU scalability, and 24 GB and 16 GB of memory, respectively.
Using a variety of independent software vendor (ISV) certifications and optimized software, the GPUs are optimized for AEC (architecture, engineering, construction) and inference workstations, providing stability and performance. Both consumer and professional Windows drivers are compatible with the Intel Arc Pro B-Series GPUs. The GPUs on Linux will be gradually enhanced with new features and optimizations, and they support a containerized software stack to make AI deployments easier. The Intel Arc Pro B-Series gives creators and AI developers scalable, affordable solutions by fusing a large memory capacity with essential software compatibility. In addition, Intel unveiled Project Battlematrix, a workstation-class Intel® Xeon®-based platform that can be customized to lower barriers for AI developers. Up to 192GB of video random-access memory and up to eight Intel Arc Pro B60 24GB GPUs are supported, allowing for medium-sized (up to 150 billion parameters) and precise AI models.
The vice president and general manager of Client Graphics at Intel, Vivian Lien, stated that the Intel Arc Pro B-Series demonstrates the company’s dedication to GPU technology and ecosystem partnerships. Small and medium-sized enterprises seeking focused solutions can now access and scale the new Arc Pro GPUs thanks to the sophisticated capabilities of the Xe2 architecture and an expanding software ecosystem. “”.
Starting in June 2025, the Intel Arc Pro B60 GPU will be sampled from add-in board partners such as ASRock, Gunnir, Lanner, Maxsun, Onix, Senao, and Sparkle, among others. In July 2025, Intel-approved resellers will begin selling the Intel Arc Pro B50 GPU.
PCIe and rack-scale deployments are supported by Intel Gaudi 3.
Intel unveiled additional deployment options for its Gaudi 3 AI accelerators as part of its AI strategy expansion.
Scalable AI inference is supported by Intel Gaudi 3 PCIe cards in current data center server environments. Because of the scalable configurations offered by AI models like Llama, clients ranging in size from small businesses to large corporations can run models ranging from Llama 3.1 8B to full-scale Llama 4 Scout or Maverick. The second half of 2025 will see the release of Intel Gaudi 3 PCIe cards.
With up to 64 accelerators per rack and 8–2 terabytes (TB) of high-bandwidth memory, the Intel Gaudi 3 rack scale system reference designs are designed for scalability and flexibility. A design that is open and modular helps prevent vendor lock-in. Additionally, these systems’ liquid cooling helps control total cost of ownership (TCO) while providing excellent performance.
With its low latency and real-time inferencing capabilities, the Intel Gaudi rack scale architecture is well-suited for executing large AI models. These setups support both custom and Open Compute Project (OCP) designs for cloud service providers (CSPs), further demonstrating Intel’s dedication to open, adaptable, and secure AI infrastructure.
The AI Assistant Builder from Intel is now accessible.
Intel AI Assistant Builder, a lightweight, open software framework for creating and executing custom AI agents locally on Intel-based AI PCs, made its debut at CES 2025 and is currently accessible in public beta release on GitHub. The Intel AI Assistant Builder enables developers and partners to rapidly develop and implement AI agents for their own companies and direct clients, as demonstrated by the new and creative solutions that Acer and ASUS are showcasing at Computex.
A booth at Computex 2025 for Intel.
Intel will exhibit its most recent advancements in processors and graphics at the Taipei Nangang Exhibition Center’s Computex 2025.
Where: J0306, Nangang Hall 1-1F.
When: May 20, Tuesday, through May 23, Friday.
Hours: 9:30 a.m. A. 5:30 p.m. A. local time in Taiwan.
about Intel.
Leading the industry in developing transformative technology that advances society and improves people’s lives is Intel (Nasdaq: INTC). In order to help our customers overcome their biggest obstacles, we are constantly striving to improve semiconductor design and manufacturing, motivated by Moore’s Law. By integrating intelligence into the cloud, network, edge, and all other computing devices, we can unleash the power of data to improve society and business. Visit Intel.com and newsroom.intel.com to find out more about Intel’s innovations.
Disclaimers and Notices.
Use, configuration, and other variables affect performance. To find out more, visit Intel.com/performanceindex.
Copyright © Intel Corporation. Intel Corporation or its subsidiaries own the trademarks Intel, the Intel logo, and other Intel marks. It is possible for others to claim ownership of other names and brands.
You can view the source version at https://www.businesswire.com. businesswire . com/news/home/20250519565090/en/.
Andrew Evangelista.
Call 1-925-487-0464.
Andrew. evangelista@intel.com.
It comes from Intel.
