ASUS is taking its ProArt creator lineup in a much more ambitious direction at IFA 2026, introducing new ProArt P16 and P14 laptops alongside the ProArt GR1X Mini PC, all built around NVIDIA’s RTX Spark superchip. The pitch is simple, even if the hardware behind it is not: bring workstation-scale local AI, high-end creative tools, and always-on AI agents to machines that creators can actually carry or keep on a desk.

For years, powerful local AI work has been largely reserved for big desktops with expensive GPUs, plenty of cooling, and the sort of space commitment that does not travel well. ASUS is betting that the next stage of the AI PC will look less like a traditional workstation and more like a premium creator laptop or a palm-sized desktop – with enough memory and compute to run demanding models without sending every prompt, file, or image to the cloud.

At the center of these new systems is NVIDIA RTX Spark, a platform that pairs an NVIDIA Blackwell RTX GPU with an NVIDIA Grace CPU through NVIDIA’s NVLink-C2C interconnect. ASUS says the chip offers up to 1 petaflop of AI performance and as much as 128GB of high-bandwidth unified memory, which is the crucial specification here. AI tasks often run out of memory long before raw processing power becomes the problem, especially when working with larger language models, high-resolution image generation, or complex 3D assets.

ASUS says RTX Spark-equipped ProArt systems can work with local language models of up to 120 billion parameters, render 3D scenes larger than 90GB, and produce 4K AI-generated video. Those claims will need real-world testing, especially once shipping software, thermals, and sustained performance are factored in. Still, the direction is notable: instead of treating AI as a small assistant feature bolted onto a conventional laptop, ASUS is positioning these machines as personal AI workstations.

ProArt P16 and P14

The new ProArt P16 and P14 are the more interesting products for creators who do not want their AI hardware tied to one desk. Both machines aim to combine RTX Spark performance with the thin, cleanly designed chassis that has become part of the ProArt identity.

The 16-inch ProArt P16 measures 12.9mm thick and weighs 1.77kg, or roughly 3.9 pounds. ASUS says that makes it its thinnest 16-inch ProArt laptop to date, as well as 14 percent thinner and 9 percent lighter than the prior generation. It also carries a 99Wh battery, which is essentially the largest capacity generally allowed on passenger flights in the U.S. without special approval.

The smaller ProArt P14 weighs 1.48kg, around 3.3 pounds, and is 13.9mm thick. It uses a 90Wh battery, so ASUS is clearly trying to avoid the usual trap of powerful mobile hardware: delivering impressive specifications that only make sense near a wall outlet. Battery-life results will depend heavily on the model and workload, of course, particularly when local AI generation or GPU rendering enters the picture.

Both laptops are aimed at the kind of creative professional whose work jumps between editing timelines, organizing media libraries, working with 3D scenes, prototyping with generative AI, and occasionally running development tools. ASUS is also emphasizing a comprehensive port selection, a haptic touchpad, and two finishes – Nano Black and Neo White – rather than forcing buyers to choose between performance and a polished, portable design.

The display specifications are similarly creator-minded. The P16 can be configured with a 4K, 120Hz variable-refresh-rate OLED panel with NVIDIA G-SYNC, while the P14 reaches up to 3K resolution. ASUS says the Lumina Pro OLED displays offer Delta E below 1 color accuracy, peak brightness up to 1,600 nits, and anti-reflection treatment. On paper, that puts the new models in serious photo, video, and design territory rather than simply being AI-focused laptops with a ProArt badge.

Local AI, without meters

The larger story ASUS is selling is not just faster image generation. It is ownership and control of the AI workflow.

Cloud AI tools are undeniably convenient, but heavy users quickly encounter monthly plan limits, per-token charges, upload concerns, and feature availability that can change without notice. A local system with enough performance to run models directly gives creators a different set of tradeoffs: the upfront hardware cost is higher, but experiments can be repeated without counting credits, sensitive assets can stay on-device, and workflows can remain available without an internet connection. ASUS specifically frames RTX Spark as a way to turn variable cloud spending into fixed personal computing capacity.

That does not mean the cloud goes away. Large frontier models and unusually demanding jobs will remain better suited to remote infrastructure for many people. But the appeal of running a capable image model, media-organizing tool, coding assistant, or specialized language model locally is becoming clearer as creative work increasingly involves private client material and large project files.

ASUS plans to preload creator software built around that local approach. MuseTree will support local generative-image creation using FLUX and WAN models, while StoryCube is designed to recognize scenes, organize media, and manage assets with AI assistance in Windows. The company is also highlighting work with ComfyUI, the widely used open-source node-based generative AI platform, as it tries to make multi-step, agent-assisted creative workflows more approachable.

The “agentic” label is becoming unavoidable across the AI industry, and it can mean almost anything in a product launch. ASUS’s practical interpretation is a chain of AI tools that can move work from one stage to another – helping generate an asset, organize it, edit it, and prepare it for the next step – rather than operating as disconnected chat boxes. Whether that feels genuinely useful will come down to the quality of the integrations, the reliability of the models, and whether creators can stay in control when the software makes decisions on their behalf.

ProArt GR1X Mini PC

If the P16 and P14 are built for mobility, the ProArt GR1X is the version of this idea for people who want local AI power running continuously at home, in a studio, or on a small office desk.

The Mini PC measures 150 x 150 x 51mm, making it small enough to disappear beside a monitor, yet ASUS is positioning it as a compact workstation-class machine for AI development, content generation, 3D rendering, and other sustained workloads. That “always on” focus matters. A compact machine can serve as a dedicated local AI box, keeping a model or workflow available while a creator edits on another computer or uses a laptop elsewhere.

Cooling is one of the obvious questions whenever serious compute is placed in a tiny enclosure. ASUS says the GR1X uses a dual-fan thermal system with 218 fan blades, up to 1.6 times greater thermal coverage, and seven levels of fan control. The company says the system is designed for sustained 24/7 operation, though independent testing will be needed to judge noise levels and long-term performance under continuous AI workloads.

Connectivity is more generous than its size suggests. The GR1X includes 10GbE Ethernet, Wi-Fi 7, Bluetooth 5.4, and support for up to four 4K displays. That makes it potentially useful beyond AI experimentation: a video-production workstation, a compact render machine, a multi-display editing setup, or a local development box that does not occupy much physical space.

ASUS will offer the GR1X in black and silver, continuing the understated ProArt design language rather than turning the device into another aggressively styled performance PC. For the intended buyer, that is probably the right call. The GR1X is meant to fit into a studio or office, not dominate it.

A different kind of AI PC

The ProArt P16, P14, and GR1X arrive as the definition of an AI PC is expanding. Early AI PCs were mostly sold on features such as background blur, live captions, webcam effects, and small on-device assistants. Those conveniences can be useful, but they do not fundamentally change what a creative professional can do with a computer.

ASUS and NVIDIA are pursuing something more substantial here: systems meant to make local AI a core part of creative and development work. The combination of Blackwell-based graphics, Grace CPU architecture, and up to 128GB of unified memory could make these devices unusually flexible for users who have outgrown lightweight AI features but do not want to build or maintain a full desktop workstation.

There are still important unanswered questions, including U.S. pricing, full availability details, software maturity, and how the laptops perform when pushed hard for long periods. Those details will determine whether RTX Spark becomes a practical new category or remains a highly capable niche for developers and well-funded production teams.

But ASUS has identified a real opening. Creators are being asked to work with bigger files, more AI tools, and increasingly complex workflows, while also keeping sensitive projects private and moving between locations. A thin 16-inch laptop, a lighter 14-inch option, and a compact desktop that can run local AI around the clock are three distinct answers to that same growing problem.