News from manufacturers

NVIDIA to Acquire Hugging Face

NVIDIA has entered into an agreement to acquire AI collaboration platform Hugging Face for approximately $12.93 billion. The service will continue to operate as an open platform supporting multi-cloud environments and non-NVIDIA computing hardware.

“NVIDIA has agreed to acquire Hugging Face for $12,930,300,000.” — NVIDIA

Hot Chips 2026: The CPU’s next chapter is being built on Arm

At Hot Chips 2026, several vendors presented upcoming processors built on the Arm architecture for enterprise and AI infrastructure. Implementations included Fujitsu's 144-core MONAKA chip, Nvidia's Vera processor utilizing custom Olympus cores, and a future IBM design engineered to run both IBM Z and Arm AArch64 code on the same CPU core.

“The company’s future Z and LinuxONE processor will natively execute both IBM Z and Arm AArch64 software on the same CPU cores.” — Arm

NVIDIA and CrowdStrike Strengthen Agentic Cybersecurity Frontier

CrowdStrike and NVIDIA have introduced SafeMind, an automated cybersecurity system integrated into the Falcon platform. The architecture uses NVIDIA Nemotron language models, specifically Nemotron 3 Ultra and Nemotron 3 Super, customized with threat data to drive offensive and defensive analysis agents.

“NVIDIA Nemotron 3 Ultra orchestrates the defensive agent harness. A fine-tuned Nemotron 3 Super powers SafeMind’s rule-generation sub-agent.” — NVIDIA

Top Arm-based innovations from August 2026 shaping the future of computing

Arm highlighted several developments across mobile and physical AI workloads, including model compression techniques for handheld devices and local processing for autonomous systems. The company also noted the launch of Google's Pixel 11 series powered by the Tensor G6 processor, alongside revenue milestones for Arm-based server hardware in AI infrastructure.

“Tensor G6 delivers 25 percent faster web browsing and 15 percent quicker app launches, while 50 percent more TPU compute.” — Arm

AMD, Cisco and HUMAIN Expand Saudi Arabia’s AI Infrastructure as AMD Instinct Systems Go Live

AMD, Cisco, and HUMAIN have deployed AMD Instinct systems to build out artificial intelligence infrastructure in Saudi Arabia. The collaboration marks the operational launch of these accelerated computing platforms within the region.

Related: AMD Instinct MI250X , AMD Instinct MI355X , AMD Instinct MI350P , AMD Instinct MI300X , AMD Instinct MI250 , AMD Instinct MI100 , AMD Instinct MI300A , AMD Instinct MI210

Exciting updates for Sneaky Sasquatch come to Apple Arcade

Apple announced a crossover event running through October 7 that introduces themes from Sneaky Sasquatch into Subway Surfers+. Both video games are available as part of the Apple Arcade catalog.

GeForce NOW Gives Gamers More Ways to Play at Gamescom 2026

NVIDIA announced several updates to its GeForce NOW cloud gaming service, including user-configurable DLSS 4.5 settings for Ultimate tier subscribers. The platform is also adding official compatibility for the Firefox web browser with streaming up to 1440p at 120 FPS, GOG single sign-on integration, and upcoming support for legacy Steam Machine hardware.

“Ultimate members can enjoy GeForce RTX-powered gaming at up to 1440p and 120 frames per second.” — NVIDIA

Related: NVIDIA GeForce RTX 3090 Ti , NVIDIA GeForce RTX 4070 Ti , NVIDIA GeForce RTX 4090 Mobile/Laptop , NVIDIA GeForce RTX 2080 Ti , NVIDIA GeForce RTX 3080 Ti , NVIDIA GeForce RTX 2080 Max-Q , NVIDIA GeForce RTX 2070 Super Max-Q , NVIDIA GeForce RTX 5060 Ti

NVIDIA NVLink Fusion Expands With NVHBM Custom High-Bandwidth Memory

NVIDIA has introduced NVHBM under its NVLink Fusion initiative, moving memory controllers into the HBM base die rather than host compute chips. This architecture is designed to decrease power usage and conserve die space on custom processors. Amazon's Annapurna Labs will adopt the technology starting with its upcoming Trainium4 accelerators.

“NVHBM, built on the same technology that NVIDIA will use for future GPUs, integrates NVIDIA’s custom memory controller into the HBM base die.” — NVIDIA

Intel at AI Infra Summit 2026

Intel announced its participation in the AI Infra Summit 2026 to present updates regarding its artificial intelligence infrastructure across cloud, enterprise, and edge domains. As part of the event, Intel CEO Lip-Bu Tan is scheduled to take part in a discussion session on September 15.

Building the systems behind the next generation of intelligent robots

Autonomous robots require compute platforms capable of managing real-time motor control, safety isolation, and high-throughput artificial intelligence inference simultaneously. Efficient operation depends on coordinating tasks between central processors and hardware accelerators without introducing latency bottlenecks. System-level integration remains critical as robotic workloads grow more demanding.

“According to the International Federation of Robotics, the global market value of industrial robot installations has reached a record USD $16.7 billion” — Arm