NVIDIA has announced the availability of RTX Spark laptops, a new platform that combines a Blackwell RTX GPU and Grace CPU in a single superchip, with up to 128 GB of unified memory and up to 1 petaflop of AI performance (FP4 precision).
Models are on sale starting around $2,599 from manufacturers including ASUS (ProArt P14 and P16), Dell (XPS 16), HP (OmniBook Ultra and X 14), Lenovo (Yoga Pro 9n), Microsoft (Surface Laptop Ultra) and MSI (Prestige N16).
What changes in practice
Unified memory allows the CPU and GPU to share the same LPDDR5X RAM pool (up to 300 GB/s bandwidth), making it easier to run large language models and AI agents locally. NVIDIA highlights initial support for agents such as GitHub Copilot, Hermes, OpenClaw and Perplexity, plus TensorRT acceleration in Adobe apps like Photoshop and Premiere.
On the gaming side, there is DLSS 5 support in titles like Gears of War: E-Day and NBA 2K27, with the company promising high performance at 1440p.
Why it matters
For developers and creators, the ability to run models of up to around 120 billion parameters locally, with long contexts, represents an alternative to cloud APIs and the VRAM limitations of traditional GPUs. The platform arrives amid strong demand for local AI and greater data control.
Prices and configurations vary by region and model. Compact desktop versions based on the same technology are expected in November.
Sources
Transparency: This content was created, edited or reviewed with the aid of artificial intelligence. Information was cross-checked with public posts on X and available internet sources. Consult the original sources for full context.
By GeekikiBot
