Nvidia's RTX Spark N1X launches in October for laptops and desktops — 18 or 20 CPU cores, paired with 5,120 or6,144 CUDA cores, up to 128GB of unified memory
- ID
- 21140
- Status
- summarized
- Published
- 04 Sep 2026, 12:00 AM
- Fetched
- 04 Sep 2026, 1:53 AM
- Provider
- Tom's Hardware
- Category
- technology
- Original URL
- https://www.tomshardware.com/laptops/nvidias-rtx-spark-n1x-launches-in-october-for-laptops-and-desktops-18-or-20-cpu-cores-paired-with-5-120-or-6-144-cuda-cores-up-to-128gb-of-unified-memory
- Source URL
- https://www.tomshardware.com/feeds/all
Summary
- Score
- 7.5
- Created
- 04 Sep 2026, 3:01 AM
- Tags
- Audience
- developersai_ml_learnersai_agent_userssaas_founders
What happened
Nvidia's RTX Spark N1X, launching in October for laptops and desktops, combines 18 or 20 CPU cores with 5,120 or 6,144 CUDA cores and up to 128GB of unified memory. This is an ARM-based CPU+GPU design sharing a single memory pool, meaning the GPU can access nearly the full 128GB without the VRAM bottleneck typical of discrete cards.
Why it matters
128GB of unified memory on a laptop/desktop means you could run roughly 70B-parameter models locally without offloading—something that currently requires multi-GPU rigs or cloud rentals. If you're budgeting for an AI development workstation in the next 6-12 months, wait for pricing and benchmarks before committing to a multi-RTX 5090 build, as this could change the cost math significantly.
Discussion angle
Compare the expected price-to-VRAM ratio of Spark N1X against current options (e.g., Mac Studio with 128GB unified memory, or a dual-RTX 5090 rig) for running 70B models locally—does this actually shift the recommendation for a Malaysian builder on a budget?