llamaperf

CMP 170HX vs DGX Spark

For running local LLMs · 27 reports across 9 models

Side A
CMP 170HX
Vendor
nvidia
VRAM
8GB
Memory
Discrete
Side B
DGX Spark
Vendor
nvidia
VRAM
128GB
Memory
Unified

Highest reported generation speed by model family

These maxima come from different community setups. Model size, quantization, GPU count, offloading and concurrency may differ. They do not establish which GPU is faster for the same workload. Check the comparison conditions →

ModelCMP 170HXDGX Spark
DeepSeek V4 Flashup to 285B29.0n=190.0n=10
Qwen3.8up to 2400Bn=1181.0n=3
Ling-3.0up to 124B40.9n=3
Qwen3.6up to 35B210.0n=2n=1
GLM-5.2up to 744B24.0n=2
DeepSeek V3n=1
GLM-5.1n=1
Kimi K320.0n=1
Museup to 30B37.0n=1

More comparisons