Nvidia Developing Proprietary CPU for Laptops

🔑 Enhanced Key Takeaways

•Nvidia's new laptop CPUs are codenamed N1 and N1X, with the N1X being the higher-end variant, and are expected to be formally introduced at Computex 2026.
•The N1X is rumored to feature up to 20 CPU cores (10 Cortex-X925 performance cores and 10 Cortex-A725 efficiency cores) and an integrated Blackwell 2.0 GPU with 48 Streaming Multiprocessors (6,144 CUDA cores), offering performance comparable to an RTX 5070 mobile GPU.
•These chips are designed for Windows on Arm laptops, with Microsoft, Nvidia, and Arm coordinating a 'new era of PC' campaign, signaling strong ecosystem support for Nvidia's entry into the consumer PC market.
•The N1X is reportedly built on TSMC's 3nm process and may integrate a CPU die designed by MediaTek with Nvidia's Blackwell GPU die using NVLink-C2C interconnect technology.
•Initial launch plans for N1-based devices were reportedly delayed from 2025 to Q1 2026, and then to late 2026, due to hardware and software compatibility issues.

📊 Competitor Analysis▸ Show

Feature/Category	Nvidia N1X (Rumored)	Apple M3 Pro (12-Core)	AMD Ryzen AI 9 HX 370	Intel Core Ultra 9 185H
Architecture	ARM-based (Cortex-X925/A725 cores)	ARM-based (6P+6E cores)	x86-64 (4 Zen 5 + 8 Zen 5c cores)	x86-64
CPU Cores	Up to 20 (10P+10E)	12	12	(Typically 16 cores, 6P+8E+2LP)
Integrated GPU	Blackwell 2.0 (48 SMs, 6,144 CUDA cores, RTX 5070-class mobile)	18-core GPU	Radeon 890M (16 CU RDNA 3+)	Intel Arc Graphics (Xe-LPG)
Process Node	TSMC 3nm	3nm	4nm	(Intel 4)
TDP Range	45W-80W (CPU+GPU package)	(Varies by configuration)	(Varies by configuration)	(Varies by configuration)
Memory Support	Up to 128GB LPDDR5X	Up to 150 GB/s bandwidth	LPDDR5x-8000	(DDR5/LPDDR5)
AI Performance (NPU)	Integrated AI accelerators (implied by Blackwell GPU and AI focus)	Neural Engine	50 TOPS XDNA 2 NPU	Integrated NPU (up to 11 TOPS)
Cinebench R23 Single-Core	(No direct N1X benchmarks yet)	(Higher than AMD Ryzen AI 9 HX 370)	~2,010 points	~1,800 points
Cinebench R23 Multi-Core	(No direct N1X benchmarks yet)	~22,000-24,000 (M3 Max)	~17,500-23,302	(Lower than M3 Max and Ryzen AI 9 HX 370)
Target Market	High-performance Windows on Arm laptops	High-performance MacBooks	High-performance Windows laptops	High-performance Windows laptops

🛠️ Technical Deep Dive

Nvidia N1X/N1 Series (Laptop SoCs):
- Architecture: Arm-based System-on-Chip (SoC) design.
- CPU Cores (N1X): Up to 20 cores, configured as 10 Cortex-X925 performance cores and 10 Cortex-A725 efficiency cores.
- Integrated GPU (N1X): Blackwell 2.0 architecture with 48 Streaming Multiprocessors (SMs), equivalent to 6,144 CUDA cores. A slightly cut-down N1X variant features 40 SMs (5,120 CUDA cores).
- Integrated GPU (N1): Two configurations: 20 SMs (2,560 CUDA cores) or 16 SMs (2,048 CUDA cores).
- Power Consumption: N1X models operate within a 45W to 80W power range, while N1 models target 18W to 45W. These figures cover the complete CPU and GPU package.
- Manufacturing Process: Reportedly built on TSMC's 3nm process.
- Memory: Supports up to 128 GB of LPDDR5X memory.
- Interconnect: May utilize NVLink-C2C to connect distinct CPU and GPU dies within the SoC, offering high bandwidth.
- Software Support: Expected to support the full CUDA software stack.
Nvidia Grace CPU (Server/HPC Context, relevant for Nvidia's CPU expertise):
- Architecture: High-performance Armv9 64-bit CPU cores (Neoverse V2).
- Core Count: Each Grace CPU features 72 cores. The Grace CPU Superchip combines two Grace CPUs for a total of 144 cores.
- Interconnect: Uses NVIDIA NVLink-C2C for coherent connection between two Grace CPUs in a Superchip, providing 900 GB/s bidirectional bandwidth.
- Memory: Utilizes server-class LPDDR5X memory with ECC, offering up to 1 TB/s memory bandwidth per Grace CPU Superchip.
- Fabric: Incorporates NVIDIA Scalable Coherency Fabric (SCF).
- Process Node: Built on TSMC N4 (4nm EUV) silicon fabrication process.
- I/O: Supports up to 128 lanes of PCIe Gen 5.

🔮 Future ImplicationsAI analysis grounded in cited sources

Nvidia's entry will significantly intensify competition in the Windows on Arm laptop market.

By providing a powerful Arm-based alternative with its strong GPU and AI ecosystem, Nvidia will challenge Qualcomm's current dominance, potentially accelerating Windows on Arm adoption and innovation.

The N1X's integrated Blackwell GPU and CUDA support could make it a compelling option for AI-heavy workloads and gaming on laptops.

Combining a powerful integrated GPU (RTX 5070-class) with Nvidia's established AI software stack (CUDA) directly on an Arm SoC offers a unique value proposition for on-device AI and gaming, despite potential x86 emulation challenges.

Nvidia's move into consumer CPUs is part of a broader strategy to vertically integrate its AI computing ecosystem from data centers to personal devices.

By offering its own CPUs, Nvidia aims to extend its dominance beyond GPUs and software into the entire AI infrastructure, including PCs, to target the growing on-device AI market and a forecasted $200 billion CPU market.

⏳ Timeline

2006

Nvidia acquired Stexar Company, which was developing an x86-compatible processor.

2008

Nvidia announced the Tegra series, an embedded client processor line for mobile devices with ARM cores and Nvidia graphics.

2011-01

Nvidia announced 'Project Denver,' an initiative to build high-performance ARM-based CPU cores for various computing platforms.

2014-10

The first consumer device featuring Denver CPU cores, the Google Nexus 9 tablet with a dual-core Tegra K1 SoC, was announced.

2022-08

Nvidia launched the Grace CPU, an Armv9 64-bit CPU designed for data centers and HPC applications.

2023-10

Reports emerged that Nvidia was developing Arm-based CPUs capable of running the Windows OS.

2025-06

An 'NVIDIA N1x' processor appeared on Geekbench, running on an HP prototype notebook with Ubuntu.

2026-05-29

Nvidia, Microsoft, and Arm posted coordinated teasers hinting at a joint announcement at Computex 2026 regarding Nvidia's Arm-based laptop chips.

Nvidia Developing Proprietary CPU for Laptops

⚡ 30-Second TL;DR

Key Points

🧠 Deep Insight

🔑 Enhanced Key Takeaways

🛠️ Technical Deep Dive

🔮 Future ImplicationsAI analysis grounded in cited sources

⏳ Timeline

📎 Sources (24)

👉Related Updates