Unverified50% confidenceBenchmarkExact time
On the DGX Spark, the public runtime is about 20 GB unpacked and the Flash weights are about 18 GiB; Flash uses 17.75 GiB of loaded CUDA tensors, causing a peak drop of about 36.06 GiB in available host memory.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/16/2026
First Seen
Valid until: 10/30/2026
Sources
Related Entities
Related Claims
UnverifiedA DGX Spark has only 128 GB of memory and cannot run the deepseek-v4-flash-0731 model that CrofAI claimed to run on it.73% similarVerifiedThe combined unified memory of two DGX Sparks totals approximately 256 GB.72% similarUnverifiedAdding weights of 39.8 GiB, KV Cache of 39.1 GiB, and activations and runtime of about 4.2 GiB gives a total of about 83.1 GiB, which will cause OOM on an 80GB GPU.72% similarUnverifiedThe DGX Spark is equipped with up to 128 GB of unified memory, AI compute performance of up to 1 petaFLOPS (FP4 precision), and is priced at approximately $3,000.66% similarVerifiedCompared to the previous generation Flash, V4.1 Flash's high-bandwidth memory requirements have been reduced to 1/4, and solid-state storage requirements have been reduced to 1/8.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1048441API
curl https://kongchang.com/api/v1/knowledge/claims/1048441MCP
get_claim(id=1048441)