llamaperf

M5 Max 128GB vs RTX A6000 48GB

For running local LLMs · 7 reports across 3 models

Side A
M5 Max 128GB
Vendor
apple
VRAM
128GB
Memory
Unified
Side B
RTX A6000 48GB
Vendor
nvidia
VRAM
48GB
Memory
Discrete

Tokens per second by model

ModelM5 Max 128GBRTX A6000 48GB
Gemma 4up to 31B7.5n=316.9n=1
Qwen3.6up to 35B5.5n=2
Qwen3.5up to 122Bn=1

More comparisons