Intel’s Next-Gen Jaguar Shores AI GPUs and Arc for Edge AI: What’s Coming in 2025

Intel has just unveiled exciting updates to its GPU lineup at the Intel AI Summit 2025 in Seoul, focusing on next-gen Jaguar Shores ‘Gaudi’ GPUs and new Arc GPUs designed specifically for Edge AI applications.

These innovations highlight Intel’s growing commitment to AI-powered hardware, promising cutting-edge performance for both inference workloads and gaming.

The big news was Intel’s partnership with SK hynix to integrate their advanced HBM4 memory into the new Jaguar Shores AI accelerators. These accelerators, built on Intel’s 18A process node, will leverage state-of-the-art RibbonFET and Backside Power Delivery technologies, significantly improving transistor density and power efficiency. Jaguar Shores is set to replace the Falcon Shores architecture, which was a failed attempt at combining x86 CPUs and GPUs into one solution. Additionally, Jaguar Shores will work seamlessly with next-gen Xeon CPUs, like Granite Rapids or the upcoming Diamond Rapids, which are expected to launch next year.

On the gaming side, Intel is making strides with its Arc series, which already includes the Battlemage-powered B-series and Pro B-series products. New Arc GPUs are slated for release in Q4 2025, with speculation surrounding the highly anticipated BMG-G31 chip that could power both AI and gaming-focused products. While specifics are scarce, Intel’s expansion into Edge AI hardware positions the company to compete head-to-head with NVIDIA and AMD, which are expected to launch their own AI solutions next year.

The collaboration with SK hynix for HBM4 memory in Jaguar Shores GPUs promises faster data throughput and better overall performance for AI servers. These new GPUs are not only expected to deliver enhanced AI processing but also to push the boundaries of what’s possible with next-gen hardware technologies. As Intel continues to evolve its Arc and Gaudi product families, enthusiasts are eager to see how these innovations will shape the future of AI and gaming hardware.

Related posts

Samsung Foundry Targets NVIDIA’s Next-Gen GPUs with 2nm Process

US Lifts Restrictions on Semiconductor Design Software Exports to China

Lisa Su’s $33 Million Stock Award Highlights AMD’s Success in the AI and GPU Markets