What can the highest-end RTX Spark configuration do?

The configuration with 128 GB of unified memory can run 120 billion parameter local models with up to one million tokens of context using local agents.

Who is RTX Spark aimed at?

The 128 GB configuration is aimed at AI enthusiasts and professionals, while configurations with less than 128 GB of RAM are aimed at creators, developers, and other traditional workstation users.

Is local AI easier than cloud-based AI?

Local AI is harder to get up and running than cloud-based AI, and RTX Spark makes a broader range of local AI models available but also increases complexity.

Large Language ModelsTop Companies' AI MovesAI Business & IndustryTop Companies AI6m agoOct 17, 2026, 06:30 JST

Microsoft, Nvidia push RTX Spark for local AI

Send on LINE

3 Key Points

What happened
Microsoft and Nvidia are promoting RTX Spark, a new class of Windows PCs for local AI, with a top configuration running 120 billion parameter models and one million tokens of context.
Why it matters
This high-end setup may suit AI enthusiasts and professionals who want the most powerful local AI capabilities in Windows 11, while lower-RAM configs target creators and developers.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Get news like this every morning, free

Context & Analysis

RTX Spark arrives as the third generation of AI PC platforms, following the original AI PCs and Copilot+ PCs, each a superset of its predecessor. Microsoft and Nvidia are marketing it for local AI, creators, developers, and gaming.

Local AI remains harder to set up than cloud-based alternatives, and while RTX Spark offers a broader range of local AI models, it also adds complexity. The three RTX Spark machines in-house for launch include the Lenovo Pro 9n 15 with a lower-end N1X processor, 18-core CPU, 5,120 core GPU, and 32 GB of RAM; the HP OmniBook Ultra 16 with a higher-end N1X, 20-core CPU, 6,144-core GPU, and 32 GB of RAM; and the ASUS ProArt P16 with the higher-end N1X and the full 128 GB of unified memory.

RTX Spark configurations should be solid for gaming, with some exceptions until developers catch up. The 128 GB configuration is seen as uniquely suited for the most demanding local workloads, and further coverage is planned over the next month.

FAQ

What can the highest-end RTX Spark configuration do?: The configuration with 128 GB of unified memory can run 120 billion parameter local models with up to one million tokens of context using local agents.
Who is RTX Spark aimed at?: The 128 GB configuration is aimed at AI enthusiasts and professionals, while configurations with less than 128 GB of RAM are aimed at creators, developers, and other traditional workstation users.
Is local AI easier than cloud-based AI?: Local AI is harder to get up and running than cloud-based AI, and RTX Spark makes a broader range of local AI models available but also increases complexity.

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.