Verified80% confidenceFactTime unknown
The CPU and GPU of DGX Spark share up to 128 GB of LPDDR5X memory, achieving memory-coherent access via NVLink-C2C high-speed interconnect.
4
Sources
80%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Gemma 4 + DGX Spark：开源大模型与桌面级算力如何改变AI开发
twitterNVIDIAAIDev
Related Entities
Related Claims
UnverifiedThe DGX Spark (GB10) has 128 GB of unified memory and can run Nemotron 3.5 Super VL using the official NVFP4 + vLLM + LiLiCorr drafter.74% similarUnverifiedRTX Spark platform laptops typically use a unified memory architecture shared between CPU and GPU (similar to Apple's Unified Memory).73% similarVerifiedNVLink-C2C technology enables high-speed interconnection between CPU and GPU.72% similarVerifiedThe DGX Spark is equipped with the GB10 Grace Blackwell Superchip, 128 GB of unified memory (LPDDR5x), and officially delivers up to 1 PetaFLOP (FP4 precision) of AI compute.71% similarUnverifiedThe DGX Spark's Grace Blackwell architecture uses a unified memory pool shared between CPU and GPU, so model weights do not need to be copied between system memory and VRAM, but two large models cannot reside in memory simultaneously.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/25938API
curl https://kongchang.com/api/v1/knowledge/claims/25938MCP
get_claim(id=25938)