Latest news with #ArcProB60


Techday NZ
21-05-2025
- Business
- Techday NZ
Intel unveils new Arc Pro GPUs & Gaudi 3 accelerators for AI
Intel has introduced new GPUs for professionals and AI developers and announced wider availability for its Gaudi 3 AI accelerators. The company unveiled the Intel Arc Pro B60 and Arc Pro B50 GPUs, expanding the Arc Pro product family for AI inference and professional workstation use. These new graphics cards feature Intel's Xe2 architecture, combining Xe Matrix Extensions (XMX) AI cores and advanced ray tracing units. According to Intel, these specifications are aimed at creators, engineers and AI developers who require high memory and scalable performance for demanding workloads. The Arc Pro B60 and Arc Pro B50 GPUs are designed to address both architecture, engineering and construction (AEC) demands as well as AI inference tasks. The new GPUs offer 24GB and 16GB of memory respectively and allow for multi-GPU configurations. Intel stated that the Arc Pro B-Series is built to offer stability and performance through independent software vendor (ISV) certifications and supports consumer and professional drivers on Windows. On Linux platforms, the Arc Pro B-Series GPUs incorporate a containerised software stack that simplifies AI deployment. Intel indicated that this stack will be enhanced over time with further features and optimisations, allowing for broader compatibility and performance improvements. The combination of higher memory capacity and targeted software support is aimed at delivering a scalable, cost-efficient computing platform for professionals working with AI and advanced workstation applications. Intel also revealed a new platform code-named Project Battlematrix, a workstation-class Xeon-based system supporting up to eight Arc Pro B60 24GB GPUs. This configuration provides up to 192GB of video RAM and is designed to support medium-sized AI models with up to 150 billion parameters. Intel asserts that this solution is intended to reduce typical friction points experienced by AI developers working on sizeable model deployments. Vivian Lien, Vice President and General Manager of Client Graphics at Intel, commented: "The Intel Arc Pro B-Series showcases Intel's commitment in GPU technology and ecosystem partnerships. 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." Intel plans to begin sampling the Arc Pro B60 GPU through add-in board partners such as ASRock, Gunnir, Lanner, Maxsun, Onix, Senao and Sparkle starting from June 2025. The Arc Pro B50 GPU will be available from Intel-authorised resellers from July 2025. In addition to the GPU announcements, Intel introduced expanded deployment options for its Gaudi 3 AI accelerators. These will be available both as PCIe cards and in rack-scale system reference designs, with the PCIe cards designed to provide scalable AI inferencing within existing data centre infrastructure. Intel stated that these cards will allow organisations from small businesses to large enterprises to run a range of models, from Llama 3.1 8B to larger deployments such as full-scale Llama 4 Scout or Maverick models, thanks to various scalable configurations. The Gaudi 3 PCIe cards are scheduled for availability in the second half of 2025. The rack-scale reference designs for Gaudi 3 support up to 64 accelerators per rack, with 8.2TB of high-bandwidth memory. The system employs an open, modular architecture designed to help organisations avoid vendor lock-in and simplify system maintenance, including liquid cooling to manage thermal performance and lower total cost of ownership. For cloud service providers, Intel noted that Gaudi 3's rack-scale approach supports both custom and Open Compute Project (OCP) infrastructures, offering flexibility and support for large-scale model training and real-time inference. The announcements coincide with Intel's 40th anniversary of operations in Taiwan. Lip-Bu Tan, Intel's Chief Executive Officer, said: "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. 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." Intel also announced the general availability of Intel AI Assistant Builder, a lightweight open framework for developing and deploying AI agents on Intel-based systems. The software, now available on GitHub, is aimed at allowing developers and partners to quickly build and roll out purpose-built AI agents, with recent examples incorporated by Acer and ASUS. The AI Assistant Builder is designed for use on local AI-capable PCs and, according to Intel, can be customised for organisational requirements and direct customer deployments. Intel stated that it will present its latest processor and graphics technologies at Computex 2025 in Taipei, including the new Arc Pro and Gaudi 3 products.


Hindustan Times
20-05-2025
- Business
- Hindustan Times
Intel launches new AI-centric GPUs at Computex 2025
Intel, at the ongoing Computex 2025 event, has unveiled a brand new range of GPUs, primarily intended for AI and workstation-related use cases. These include new Intel Arc Pro GPUs as well as the Intel Gaudi 3 AI accelerators. These are professional-grade GPUs, primarily designed for AI workloads and to be used in work-centre environments. Alongside this, Intel has also launched its AI Assistant Builder, which happens to be open source. This is a framework for building and running AI agents on Intel-based AI PCs. It has been released on GitHub in a beta capacity. Here's what you need to know about the latest announcements. Also Read: Paytm introduces option to hide specific transactions from history: Here's how it works Intel has launched the Arc Pro B60 and Pro B50 GPUs, which are based on the Xe2 architecture. These feature Intel's Xe Matrix Extensions (XMX) AI cores and ray tracing units, enabling high-end performance for creators, developers, and engineers. Intel says that both GPUs are designed for today's demanding AI inference workloads and workstation applications. The Arc Pro B60 GPU comes with 24GB of memory, while the Arc Pro B50 GPU offers 16GB of memory. Both also support multi-GPU scalability. Intel also states that these GPUs are optimised for AEC (Architecture, Engineering, and Construction) and inference workloads. These GPUs are compatible with both consumer and professional drivers on Windows. On Linux, they support a containerised software stack to simplify AI deployments. Intel also announced new deployment options for its Gaudi 3 AI accelerators. The Gaudi 3 PCIe cards now support AI inferencing within existing data centre server environments. According to Intel, customers running AI models like Llama can benefit from the ability to run models ranging from Llama 3.1 8B to full-scale models like Llama 4 Scout or Maverick, thanks to the scalability of Intel's hardware. Intel says its Gaudi 3 Rack Scale System reference designs are built for scalability and flexibility, supporting up to 64 accelerators per rack and 8.2TB of bandwidth memory. The systems also feature a modular design to prevent vendor lock-in and utilise liquid cooling to ensure performance while keeping total cost of ownership low. Intel states that the Arc Pro B60 GPU will be sampled by board partners such as ASRock, Sparkle, Gunnir and others starting June 2025. The Arc Pro B50 GPU will be available from Intel-authorised resellers in July 2025, with broader availability expected in the second half of 2025. MOBILE FINDER: iPhone 16 LATEST Price, Specs, And More


Time of India
20-05-2025
- Business
- Time of India
Intel taps AI creators, developers with new Arc Pro GPUs
NEW DELHI: Intel at the Computex 2025 event on Tuesday launched two new graphics processors under its Arc line-up, including the Arc Pro B60 and the Arc Pro B50, to support artificial intelligence (AI) inferencing and professional workloads. The new Intel Arc Pro GPUs , based on the Xe2 architecture, feature Intel Xe Matrix Extensions (XMX) AI cores and advanced ray tracing units, extending high-performance capabilities to creators, developers, and engineers. The move also signals Intel's strategy to actively tap into the AI boom, which has boosted revenues of its rivals, Nvidia and Advanced Micro Devices (AMD). Intel said its latest GPUs are 'designed for demanding AI inference workloads and workstation applications', with AI-ready features, 24GB (Arc Pro B60) and 16GB (Arc Pro B50) memory, and multi-GPU scalability. The GPUs are optimised for AEC (architecture, engineering, construction) and inference workstations, offering stability and performance through a wide range of independent software vendor (ISV) certifications and optimised software. Intel also revealed a configurable workstation-class Intel Xeon-based platform – code-named Project Battlematri – designed to reduce AI developer friction points. It supports up to eight Intel Arc Pro B60 24GB GPUs to enable medium-size (up to 150 billion parameters) and accurate AI models with up to 192GB of video random-access memory, the chipmaker said in a statement. 'The Intel Arc Pro B-Series showcases Intel's commitment in GPU technology and ecosystem partnerships,' said Vivian Lien, vice president and general manager (client graphics), 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.' This product launches coincide with the 40th anniversary of Intel in Taiwan. '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, 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.' Intel said its Gaudi 3 AI Accelerators are now available in PCIe and rack-scale systems, offering scalable, open solutions for enterprise and cloud AI inferencing . It added that the Intel AI Assistant Builder is now publicly available on GitHub, enabling developers to create local, purpose-built AI agents optimised for Intel platforms.
Yahoo
19-05-2025
- Business
- Yahoo
Intel debuts AI GPUs for workstation system as it works to gain ground on Nvidia, AMD
Intel (INTC) is expanding its graphics chip offerings with the debut of its Arc Pro B50 and B60 series workstation cards. The options, starting at $299 and $500 and higher, respectively, are designed for customers who don't necessarily need or want to use cloud-based AI processors from the likes of Amazon (AMZN), Google (GOOG, GOOGL), or Microsoft (MSFT). Workstation systems provide strong security since data is processed on-site rather than being sent to the cloud. They're also generally more customizable and offer instant access to powerful graphics processing units (GPUs) without the potential for having to wait for availability via cloud services. Intel has been working on its Arc GPU platform for years. It currently offers its B-series chips for gamers, dubbed Battlemage, and an A-series range for workstations, though those cards are meant more for enterprise graphics and visual capabilities rather than AI. The Arc Pro B50 and B60, however, offer more memory and functionalities that are designed specifically with AI in mind. The B50 is a mainstream chip that Intel says is designed for AI-assisted software engineering. According to the company, entry workstation GPUs make up about 30% of the workstation market, offering Intel a clear opportunity to grab market share. "We expect this to be very competitive and a very attractive solution, particularly for design and engineering," Intel vice president and general manager of client graphics Vivian Lien told Yahoo Finance. To do that, Intel has equipped the Arc Pro B50 with 16GB of memory and support for up to 170 TOPS, or trillions of operations per second. TOPS is a generally accepted means of measuring an AI chip's rough performance capabilities. The Arc Pro B60 has 24GB of memory and is capable of up to 197 TOPS. According to Lien, that chip is aimed at workstation customers who want to run resource-intensive AI applications. Intel, like Nvidia, is touting the need for customers to opt for high-end graphics chips to power AI models into the future. To that end, the company is also introducing its Project Battlematrix workstation system. A complete PC, Project Battlematrix combines two to eight Arc Pro B60 GPUs supporting up to 192GB of memory with an Intel Xeon processor. Customers can use it to run AI software across their businesses. Intel isn't the only company with its own workstation GPUs or heavy-duty AI desktop systems. Nvidia (NVDA) has also announced its own RTX Pro AI chips and cards, as well as two desktop systems: the DGX Spark and DGX Station. The DGX Spark is a pint-sized supercomputer that runs on Nvidia's Grace Blackwell GB10 superchip. Then there's Nvidia's DGX Station. Outfitted with the company's GB300 Grace Blackwell Ultra superchip, the DGX Station provides up to 784GB of memory for AI applications. Intel is working to prove that it's not backing away from the GPU market while also proving it has what it takes to reclaim its crown atop the CPU market. While the B50 and B60 might not be as powerful as Nvidia's offerings, they're clear indications that the company is dedicated to gaining ground in the AI workstation market. But it still has to prove it can hang in the larger, more lucrative hyperscaler market. And that remains to be seen. Email Daniel Howley at dhowley@ Follow him on X/Twitter at @DanielHowley.