NVIDIA's 'MacBook Pro' Leaked: Huang's Self-Developed CPU!
NVIDIA teases a new PC era with its upcoming N1X chip, a self-developed ARM CPU paired with a Blackwell GPU, targeting AI-native laptops. The chip features 20 cores, 6144 CUDA units, and 128GB unified memory, but gaming performance is limited.
NVIDIA has officially teased its entry into the PC market with a self-developed CPU, the N1X, marking a significant shift in the AI hardware landscape. In a cryptic post on social media, NVIDIA shared coordinates (25.0528, 121.5990) pointing to the Taipei Music Center, the venue for Computex's GTC Taipei keynote, along with the phrase "A new era of PC." This strongly suggests the imminent announcement of a laptop powered by the N1X, running Windows on Arm, and designed to challenge Apple's M-series chips in the AI domain.
The N1X chip, developed in collaboration with MediaTek and built on TSMC's N3B process, is a system-on-a-chip (SoC) featuring a 20-core ARM CPU, a Blackwell GPU with 6144 CUDA cores, and 128GB of LPDDR5X unified memory. This configuration is reminiscent of NVIDIA's DGX Spark mini PC, which uses the GB10 superchip. The 6144 CUDA cores match the desktop RTX 5070, highlighting the chip's potential for AI and compute tasks. However, the unified memory architecture means GPU bandwidth is limited to around 273 GB/s, making high-end gaming impractical. Additionally, running legacy x86 games requires emulation, further hampering gaming performance.
Instead, the N1X laptop is positioned as an AI-native device, enabling local inference and development without reliance on cloud subscriptions. NVIDIA's move aligns with a broader industry trend toward local AI compute, with companies like ASUS and Lenovo also launching AI desktop solutions. The article draws a historical parallel to the printing press, suggesting that affordable, powerful local AI hardware could democratize access to AI, much like the printing press made knowledge accessible. For creators and developers, this means one-time hardware investment and near-zero marginal cost for token generation, unlocking new possibilities for automation and prototyping.
While pricing details remain undisclosed, the "NVIDIA premium" is anticipated. The announcement is expected at Computex Taipei, where Jensen Huang will likely deliver a keynote. If successful, the N1X could redefine the PC market for AI users, but its limited gaming appeal may restrict its mainstream adoption initially.