NVIDIA's N1X SoC: A Powerful Hybrid Chip Challenging the AI PC Landscape

Nvidia N1X SoC leaks with the same number of CUDA cores as an RTX 5070 -- N1X specs align with the GB10 Superchip

Nvidia's long-rumored and long-delayed N1X SoC has broken cover once again, this time revealing its GPU capabilities through a fresh Geekbench OpenCL listing. We've seen various CPU-focused leaks surrounding this chip before, but this provides our first genuine look at the GPU. That being said, this isn't final silicon -- it's an early engineering sample -- but the details are enough to show where Nvidia is headed with its first consumer-class ARM SoC for laptops and possibly desktops. The Geekbench entry confirms a 20-core CPU setup split into two 10-core clusters, built on Nvidia's Grace architecture. More importantly, it confirms 48 Streaming Multiprocessors (SMs), translating to 6,144 CUDA cores -- the exact count as the desktop GeForce RTX 5070. Interestingly, these specs also match Nvidia's GB10 Superchip, which currently powers the DGX Spark AI mini-PCs, suggesting that N1X might be a repurposed, possibly lower-power version aimed at the mainstream market. After all, they're both ARM-based chips developed in unison with MediaTek. Performance, however, is still in the rough. This sample clocked in at a modest 1.05 GHz and scored 46,361 in OpenCL, roughly in the territory of the RTX 2050. The reason is power and frequency limits, which are typical of early prototypes, as well as the absence of dedicated GDDR memory, since the SoC relies on shared LPDDR5X. Even in this state, the N1X's iGPU is already outpacing every modern integrated GPU, including Apple's M3 Max and AMD's 890M, which top out at around 37,500 in similar benchmarks. Even in this state, the N1X's iGPU is already outpacing every modern integrated GPU, including Apple's M3 Max and AMD's 890M, which top out at around 37,500 in similar benchmarks. The GPU performance ceiling is higher, given that a fully powered RTX 5070 boosts to 2.5 GHz with a 250W TDP, while the N1X is capped at around 120W for the entire chip. This isn't the first time the N1X has appeared on Geekbench. Earlier leaks revealed its CPU performance was competitive but not earth-shattering, which is to be expected from unfinished drivers and firmware. With the latest listing, Nvidia is signaling a hybrid approach: pairing a Blackwell GPU core array with an ARM-based CPU cluster, aiming for a balance of AI performance, gaming, and efficiency -- similar to how AMD's Strix Halo and Apple's M-series chips are positioned. There's still no launch date in sight. Rumors suggest a Q1 2026 release, possibly timed to coincide with the next wave of AI-focused Windows laptops, as Microsoft appears to be holding up on that at the moment. For now, Nvidia's N1X sits in the shadows, a prototype with the specs of a mid-tier discrete GPU and ambitions to push iGPU performance into uncharted territory. If Nvidia gives this chip the power and bandwidth it needs, the N1X could be the first ARM-based SoC to seriously challenge x86 giants like AMD and Intel on high-performance laptops -- and maybe even take a bite out of Apple's lead in the premium AI laptop segment.

NVIDIA's new N1X AI PC chip rumor: same GPU core count as RTX 5070, faster than all other iGPUs

TL;DR: NVIDIA's upcoming N1X AI PC processor, featuring Arm CPU and Blackwell GPU cores, achieves impressive Geekbench OpenCL scores surpassing all integrated GPUs and rivaling the GeForce RTX 5070. Despite a lower clock and shared memory, it targets AMD's Strix Halo APU with a 2026 launch after silicon modifications. NVIDIA is still cooking its upcoming N1X AI PC processor, based on Arm CPU cores and Blackwell GPU cores. But now we've got some Geekbench scores that show it's as fast as a GeForce RTX 5070, and faster than all other integrated GPUs on the market. In some new Geekbench OpenCL browser benchmarks, we have the upcoming NVIDIA N1X processor scoring 46,361 points. We've also got some details on the Blackwell-based GPU which rocks 6144 CUDA cores and 48 SM units, the same core count as the GeForce RTX 5070 which is based on the GB205 "Blackwell" GPU. NVIDIA's GeForce RTX 5070 features the GB205 GPU, which contains 192 TMUs, 80 ROPs, 48 RT Cores, and 192 Tensor Cores. There's also 12GB of GDDR7 memory with up to 672GB/sec of memory bandwidth, a GPU boost clock of up to 2512MHz and a 250W TDP. There are some stark differences between the RTX 5070 and upcoming N1X processor, with the Geekbench result seeing the N1X barely pushing 1.05GHz (1048MHz to be exact). The N1X also doesn't have dedicated VRAM, but rather shares the LPDDR5X memory onboard the SoC itself. The TDP is also radically lower with a maximum 120W TDP. Even with the 46,361 points on the OpenCL benchmark on Geekbench, we shouldn't expect RTX 5070 performance -- but, it does beat every other integrated GPU on the market -- an impressive feat in its engineering sample state. NVIDIA's upcoming N1X processor will be aimed as a competitor to AMD's current-gen Strix Halo APU, which is a powerful APU on its own, with a fantastic RDNA 3.5-based Radeon 8060S GPU that is capable of 1080p and 1440p 120Hz+ gaming, and even light 4K 60FPS if you dial down some of the visual settings in your games. We've been reporting that NVIDIA's new N1X AI PC processor has hit some issues lately requiring some modifications performed to the silicon, and that its launch has been pushed back until 2026. We should hopefully have some more concrete news on N1X at CES 2026, which is less than 6 months away now.

NVIDIA N1X SOC Features As Many Cores As The GeForce RTX 5070 GPU, 6144 "Blackwell" Cores & Faster Than All Modern iGPUs In Engineering State

The first details of the GPU featured on the N1X SoC have started to emerge from Geekbench, which offers the same core count as RTX 5070. NVIDIA N1X SOC Packs A "Blackwell" GPU With The Same Core Count As The RTX 5070, Much Work To Be Done On The Performance Side A few hours ago, the first entry of NVIDIA's N1X showed up in Geekbench's OpenCL browser. While we have already reported the CPU aspects of the SOC, the GPU side information was still a bit lacking. But now we know what the chip has to offer. The N1X entry shows that the SOC will be packing a Blackwell GPU with the same core count as the RTX 5070. This means we will be seeing 6144 core arranged within 48 SM units. The RTX 5070 uses the GB205 GPU, but it is unclear if the N1X is using a similar silicon or a custom variant. The RTX 5070 packs a total of 192 TMUs, 80 ROPs, 48 RT Cores, and 192 Tensor Cores. It also comes with dedicated 12 GB of GDDR7 memory with up to 672 GB/s bandwidth. Clock speeds for the dedicated graphics card boost up to 2512 MHz on a 250W TDP. The Geekbench listing shows that the N1X's GPU was barely pushing it beyond 1.05 GHz, which is vastly lower than the RTX 5070. Furthermore, the chip doesn't have dedicated memory, but instead features the LPDDR5X standard, which is shared between the entire SOC. And lastly, the chip is expected to feature a maximum TDP range of up to 120W, so even if the GPU does get 100W power in graphics-intensive or AI scenarios, it is still 150W behind the dedicated part. Looking at the performance figures, the N1X scored 46,361 points, which puts it above and beyond the other modern day iGPUs but way behind the RTX 5070. Once again, this is due to the fact that not only is this chip severely constrained in the power department, but it is still an early engineering sample. NVIDIA will be tweaking the performance for its N1X SOCs massively before launch. Also, the N1X, given its 120W or so TDP, should be compared against the Strix Halo "AMD Ryzen AI Max" offerings, which feature a similarly large GPU. Those provide much better performance, but once again, it remains to be seen just how much performance NVIDIA can extract out of these chips before launch. Current rumors allege that the N1X SOC will be launching sometime in 2026, with some claiming the first half and others claiming the latter half of the year. With that said, N1X will make the SOC segment a lot more competitive than it is, and will be NVIDIA's first-ever chance to really compete against x86 and Arm giants such as Intel, AMD, Qualcomm, and Apple. NVIDIA N1X SOC Specs (Preliminary):