Intel announced a significant expansion of its professional computing and artificial intelligence (AI) portfolio with the launch of new Intel® Arc™ Pro B-Series graphics processing units (GPUs), the latest Intel® Gaudi® 3 AI accelerators, and the public release of the Intel® AI Assistant Builder. These innovations are tailored for professionals, developers, and enterprises aiming to accelerate AI workloads and creative applications with flexible, high-performance solutions.
This major product reveal coincides with a milestone moment—Intel’s 40th anniversary in Taiwan—marking four decades of collaborative innovation with one of the world’s most vital technology ecosystems. The enduring strength of the x86 architecture continues to drive progress in AI and next-gen computing experiences.
“For the past 40 years, the power of our partnership with the Taiwan ecosystem has fueled innovation that has changed our world for the better,” said Lip-Bu Tan, CEO of Intel. “This week, we are renewing our commitment to our partners as we work to build a new Intel for the future. Together, we will create great products that delight our customers and capitalize on the exciting opportunities ahead.”
Next-Gen Arc Pro GPUs Boost AI and Workstation Capabilities
Intel introduced the new Arc Pro B60 and B50 GPUs, built on its advanced Xe2 architecture. Featuring Intel® Xe Matrix Extensions (XMX) AI cores and state-of-the-art ray tracing units, these GPUs are engineered to handle demanding professional workloads including architecture, engineering, construction (AEC), and AI inferencing.
The Intel Arc Pro B60 GPU delivers 24GB of memory, while the B50 offers 16GB—both supporting multi-GPU scalability. These GPUs are backed by a robust ecosystem of ISV certifications, ensuring performance and reliability across key professional applications. Windows users benefit from both consumer and pro driver support, while Linux developers can leverage a containerized software stack for streamlined AI deployment. Progressive updates will continue to enhance performance and functionality over time.
Intel also unveiled a new AI developer-focused platform, code-named Project Battlematrix. This modular workstation-class system, built on Intel® Xeon® processors, supports up to eight Arc Pro B60 24GB GPUs—enabling training and deployment of models with up to 150 billion parameters and 192GB of video memory.
Also Read: NVIDIA Launches NVLink Fusion for Custom AI Infrastructure
“The Intel Arc Pro B-Series showcases Intel’s commitment in GPU technology and ecosystem partnerships,” said Vivian Lien, vice president and general manager of Client Graphics at Intel. “With Xe2 architecture’s advanced capabilities and a growing software ecosystem, the new Arc Pro GPUs deliver accessibility and scalability to small and medium-sized businesses that have been looking for targeted solutions.”
Sampling of the Arc Pro B60 GPU will begin in June 2025 through partners such as ASRock, Gunnir, Lanner, Maxsun, Onix, Senao, and Sparkle. The Arc Pro B50 will be available via Intel-authorized resellers starting July 2025.
Intel Gaudi 3 Now in PCIe and Rack Scale Formats for AI at Scale
Further advancing its AI roadmap, Intel introduced two new deployment options for the Intel Gaudi 3 AI accelerator:
-
The Gaudi 3 PCIe card is designed for seamless integration into existing data center infrastructure. It supports a wide range of AI models, including Llama 3.1 8B and large-scale versions of Llama 4 (Scout and Maverick). With scalable configurations, it caters to businesses of all sizes. Availability is slated for the second half of 2025.
-
The Gaudi 3 rack scale system delivers enterprise-grade performance, supporting up to 64 accelerators per rack and offering 8.2 terabytes of high-bandwidth memory. Its open, modular design ensures vendor flexibility and future scalability. Integrated liquid cooling optimizes energy efficiency and lowers total cost of ownership.
Built with low-latency inferencing and large-model workloads in mind, Gaudi 3’s rack architecture aligns with Intel’s vision for open, secure, and adaptable AI infrastructure. The systems support both custom and Open Compute Project (OCP) designs, making them an ideal choice for cloud service providers seeking flexibility and control.
AI Assistant Builder Now Open to Developers
In a move to further empower the developer community, Intel announced the public availability of the Intel® AI Assistant Builder on GitHub. This open-source tool enables the creation of lightweight, task-specific AI agents optimized for Intel hardware. With it, developers can more easily integrate AI into their workflows—accelerating innovation at the edge, on-premises, or in the cloud.
