Calculator SubmitSign in

DGX Spark vs H100 80GB

For running local LLMs · 4 reports across 4 models

Side A

Vendor: nvidia
VRAM: 128GB
Memory: Unified

Side B

Vendor: nvidia
VRAM: 80GB
Memory: Discrete

Tokens per second by model

Model	DGX Spark	H100 80GB
GLM-5.1	—n=1	—
Gemma 4up to 31B	—	125.3n=1
Qwen3.5up to 27B	—	—n=1
Qwen3.6up to 35B	—	45.0n=1

More comparisons