NVIDIA RTX Spark Windows laptops go on sale from $2,399

NVIDIA says RTX Spark laptops are now available, putting its latest local AI PC platform into slim Windows machines from major manufacturers. For creators, freelancers and small teams, the launch matters because it moves high-end AI hardware into a portable form factor rather than a desktop-only setup.

The systems start at $2,399, according to NVIDIA, and include laptops from ASUS, Dell, HP, Lenovo, Microsoft and MSI, with more models planned. The company marked the launch with retail events around the world, including a Berlin appearance by founder and CEO Jensen Huang.

What RTX Spark adds to Windows laptops

RTX Spark combines an NVIDIA Blackwell RTX GPU and NVIDIA Grace CPU in what NVIDIA describes as a single superchip. The company says the platform supports up to 1 petaflop of AI compute and up to 128GB of unified memory.

That specification is the core pitch for local AI: more work can happen on the device itself, rather than only through cloud services. NVIDIA is positioning these laptops for local AI, content creation and gaming, all in premium Windows PCs.

For Emonarc readers, the practical question is not whether every workflow needs this much hardware. It is whether a portable machine that can run heavier AI workloads locally is worth adding to a tool stack that may already include cloud-based apps for writing, image generation, video editing, coding or automation.

If your work involves testing AI models, building demos, editing media, or running repeated local experiments, the unified memory ceiling is likely to be the headline feature to examine. If most of your AI use happens inside browser-based tools, the buying decision is less obvious and should come down to whether local performance changes your daily workflow.

The first PC makers and starting price

NVIDIA says RTX Spark laptops are available from six PC brands at launch:

• ASUS
• Dell
• HP
• Lenovo
• Microsoft
• MSI

The company says pricing starts at $2,399 and that additional systems are on the way. It did not provide a full model list in the source announcement, so buyers will need to compare configurations from each PC maker directly.

The arrival of Microsoft among the named brands also fits a broader push toward AI-capable Windows PCs. Emonarc recently covered Microsoft’s Nvidia-powered Surface AI PCs and Windows 11 agent features, while NVIDIA had already set expectations with global RTX Spark launch events.

For small studios and solo operators, the vendor list is useful because it suggests RTX Spark will not be limited to a single niche machine. But the source announcement does not say how configurations differ across brands, so buyers should check memory, GPU details, ports, thermals and warranty terms before treating all RTX Spark laptops as interchangeable.

Launch events put local AI in front of creators

NVIDIA’s most detailed launch story came from Berlin, where RTX Spark appeared during the city’s Festival of Lights. The company says the Berlin TV Tower featured RTX Spark-themed lighting as part of the celebration.

At XPERION, the gaming venue inside Saturn at Alexanderplatz, attendees tried laptops from participating PC makers. NVIDIA says the demos covered gaming, content creation and local AI workflows.

Huang made a surprise appearance at the Berlin event, where he greeted attendees, handed out popcorn, took selfies and gave away five signed RTX Spark laptops, according to the company.

Berlin was one of six retail launch celebrations worldwide. NVIDIA also highlighted events in Beijing, Istanbul, Seoul, San Francisco and Taipei, with hands-on demos, creator programming, founder chats or product showcases depending on the city.

What creators and small teams should watch next

The most important follow-up will be the real-world shape of the laptop lineup. NVIDIA has given the platform-level details, but buyers still need system-level information from ASUS, Dell, HP, Lenovo, Microsoft and MSI before making a practical purchase decision.

For creators, that means looking beyond the phrase “local AI” and matching the machine to actual work: image and video production, AI-assisted coding, client demos, model testing, or on-device experimentation. For small teams, the question is whether one high-end portable AI workstation can replace a desktop setup, support client-facing demos, or reduce friction when moving between office, studio and travel.

NVIDIA says developers and AI enthusiasts can use NVIDIA-tested recipes, playbooks and other resources for running local AI on RTX Spark. That may be just as important as the hardware itself: local AI PCs are only useful if the software setup is manageable enough for working teams, not just enthusiasts.