llamaperf

CMP 170HX vs M5 Max 128GB

For running local LLMs · 20 reports across 7 models

Side A
CMP 170HX
Vendor
nvidia
VRAM
8GB
Memory
Discrete
Side B
M5 Max 128GB
Vendor
apple
VRAM
128GB
Memory
Unified

Highest reported generation speed by model family

These maxima come from different community setups. Model size, quantization, GPU count, offloading and concurrency may differ. They do not establish which GPU is faster for the same workload. Check the comparison conditions →

ModelCMP 170HXM5 Max 128GB
Qwen3.6up to 35B210.0n=279.4n=4
Qwen3.8up to 2400Bn=155.0n=5
Gemma 4up to 31B7.5n=3
DeepSeek V4 Flashup to 285B29.0n=1n=1
Qwen3.5up to 122Bn=1
StepFun 3.733.9n=1
Tencent-HY3up to 295B32.4n=1

More comparisons