Unverified50% confidenceFactExact time
On the same hardware of a DGX Spark, differences in configuration can result in performance gaps of over 40%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/23/2026
First Seen
Valid until: 12/22/2026
Sources
Related Entities
Related Claims
UnverifiedThe test device is a 128G compute module (referred to in the video as the Thor T5000) claimed to have performance twice that of the NVIDIA DGX Spark.64% similarVerifiedThe CPU and GPU of DGX Spark share up to 128 GB of LPDDR5X memory, achieving memory-coherent access via NVLink-C2C high-speed interconnect.62% similarUnverifiedThe 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.62% similarVerifiedDGX Spark is a desktop-level AI supercomputer released by NVIDIA at the 2025 GTC conference, featuring the GB10 superchip based on the Grace Blackwell architecture.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/942565API
curl https://kongchang.com/api/v1/knowledge/claims/942565MCP
get_claim(id=942565)