DGX Spark vs M5 Max 64GB

For running local LLMs · 5 reports across 3 models

Side A

Vendor: nvidia
VRAM: 128GB
Memory: Unified

Side B

Vendor: apple
VRAM: 64GB
Memory: Unified

Tokens per second by model

Model	DGX Spark	M5 Max 64GB
Qwen3.6up to 35B	—	63.0n=3
GLM-5.1	—n=1	—
Gemma 4up to 31B	—	97.0n=1

More comparisons