llamaperf

H100 80GB vs M3 Max 128GB

For running local LLMs · 2 reports across 1 model

Side A
H100 80GB
Vendor
nvidia
VRAM
80GB
Memory
Discrete
Side B
M3 Max 128GB
Vendor
apple
VRAM
128GB
Memory
Unified

Tokens per second by model

ModelH100 80GBM3 Max 128GB
Qwen3.6up to 35B45.0n=15.5n=1

More comparisons