llamaperf

M1 Max 32GB vs RTX Pro 6000 Blackwell

For running local LLMs · 5 reports across 3 models

Side A
M1 Max 32GB
Vendor
apple
VRAM
32GB
Memory
Unified
Side B
RTX Pro 6000 Blackwell
Vendor
nvidia
VRAM
96GB
Memory
Discrete

Tokens per second by model

ModelM1 Max 32GBRTX Pro 6000 Blackwell
Llama 3.3n=3
Gemma 4up to 31B10.0n=1
Qwen3.6up to 35Bn=1

More comparisons