RTX Spark Ignites Global Celebration as New Windows PCs Arrive

NVIDIA says RTX Spark laptops are now available from ASUS, Dell, HP, Lenovo, Microsoft, and MSI, starting at $2,399. The pitch is a single superchip that pairs a Blackwell RTX GPU with a Grace CPU, with up to 128GB of unified memory and up to a petaflop of AI compute, in slim Windows machines.

If that holds up in real use, it matters. A lot of meaningful local AI work, like running bigger models on-device, keeping data off the cloud, or creative workloads, has been stuck on desktops or rented GPUs. This is NVIDIA's bid to make that portable.

Let's be clear about what this post is: a vendor recap of six launch parties in Berlin, Beijing, Istanbul, Seoul, San Francisco, and Taipei. There were long lines, a TV tower lit up, and Jensen Huang handing out popcorn and five signed laptops. None of that tells us how these machines perform.

The specs come with qualifiers. Phrases like "up to 128GB" and "up to 1 petaflop" describe the top configuration and a peak figure. They don't describe the $2,399 entry model or sustained performance in a thin chassis. We'd want independent testing on thermals, battery life under AI load, and how Windows apps actually handle this hardware before calling it a new generation of anything.

Still, we're rooting for the idea. Large unified memory on a laptop is what makes local models practical. A broad set of OEMs shipping on day one suggests this isn't a one-off showcase. NVIDIA also says it's publishing tested recipes and playbooks for running local AI on the hardware, which is the kind of unglamorous support developers actually need.

The downside is price and lock-in. At this entry point, local AI on a laptop stays a premium tier for now. And the more the experience depends on NVIDIA's own stack, the more Windows local AI starts to mean NVIDIA local AI. I'm watching whether the post-launch story is about software maturity rather than event photos.