NVIDIA Enters Desktop AI Market with GB10 Grace Blackwell Chip, ASUS Leads Launch

NVIDIA's Groundbreaking Entry into Desktop AI Market

NVIDIA is making a significant move into the desktop PC chip market with its new GB10 Grace Blackwell platform, specifically designed for AI workloads. The company's first desktop PC chip is set to launch this month, with ASUS leading the charge by introducing systems based on this platform starting July 22 1

1

2

The GB10 Grace Blackwell: A Powerful AI-Focused SiP

Source: Tom's Hardware

At the heart of this innovation is NVIDIA's GB10 Superchip system-in-package (SiP). This cutting-edge chip integrates:

1. A Grace CPU featuring 10 high-performance Arm Cortex-X925 cores (up to 3.90 GHz) and 10 power-efficient Cortex-A725 cores.
2. A Blackwell GPU capable of delivering 1 PetaFLOPS of FP4 compute throughput for AI workloads 1

   1

   .

The GB10 boasts a unified memory architecture with a 256-bit memory interface supporting 128GB of LPDDR5X memory, achieving bandwidths up to 273 GB/s. This memory system is comparable to Apple's M4 Pro, highlighting its capability for handling large, memory-intensive AI models efficiently 1

1

2

ASUS Ascend GX10: Showcasing the GB10 Platform

Source: TweakTown

ASUS is taking the lead in bringing this technology to market with its Ascend GX10 mini-PC. The system is positioned as an AI solution that delivers data center-level performance in a compact form-factor suitable for workstations and edge deployments 1

1

3

Key features of the ASUS Ascend GX10 include:

• Powered by NVIDIA's GB200 Grace Blackwell Superchip
• 128GB of LPDDR5X memory
• PCIe Gen5 connectivity
• NVMe SSD
• Custom cooling solution 3

  3

Market Positioning and Competition

NVIDIA is positioning the GB10 platform as an ideal solution for building and running Large Language Models (LLMs), generative AI applications, and other demanding workloads on desktop-class systems. The unified memory architecture, massive FP4 throughput, and strong single-thread performance are highlighted as key advantages over traditional CPU-GPU setups 1

1

2

The ASUS Ascend GX10 is expected to compete with systems powered by AMD's new Ryzen AI Max 300 series "Strix Halo" APUs and Apple's Mac Studio. It's anticipated to be priced similarly to NVIDIA's own DGX Spark small form-factor system, which costs around $3,000 1

1

3

Related Stories

ASUS Unveils Ascent GX10: A Compact AI Supercomputer Powered by NVIDIA's GB10 Grace Blackwell Superchip

ASUS Unveils ExpertCenter Pro ET900N G3: A Desktop Powerhouse with NVIDIA's GB300 Blackwell Ultra Superchip

NVIDIA's DGX Spark: A Compact AI Powerhouse Set to Launch in July 2025

NVIDIA Unveils GB10 Superchip: A Leap Forward in AI PC Technology

Industry Impact and Future Prospects

This launch marks NVIDIA's first serious foray into the desktop CPU business. While not a traditional x86-based processor, the Grace CPU's integration into desktop systems opens new possibilities for AI developers and researchers needing edge AI compute without relying on conventional x86 platforms 3

3

Other major PC manufacturers, including Dell, HP, and Lenovo, are also preparing their versions of DGX Spark systems based on the same NVIDIA platform, indicating a broader industry shift towards AI-focused desktop workstations 1

1

2

Performance Expectations

While leaked Geekbench figures suggest the GB10's general compute power is comparable to Qualcomm's Snapdragon X Elite and Apple's M3 processor, it's important to note that the chip's primary focus is on AI workloads. The true performance benefits are expected to be seen in tasks leveraging the GPU compute capabilities and unified memory architecture 1

1

2

As the launch date approaches, the tech industry eagerly anticipates the real-world performance and impact of NVIDIA's GB10 Grace Blackwell platform on the desktop AI market.