llamaperf

M5 Pro 64GB vs RTX 5070 Ti

For running local LLMs · 20 reports across 4 models

Side A
M5 Pro 64GB
Vendor
apple
VRAM
64GB
Memory
Unified
Side B
RTX 5070 Ti
Vendor
nvidia
VRAM
16GB
Memory
Discrete

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 →

ModelM5 Pro 64GBRTX 5070 Ti
Qwen3.8up to 2400B27.6n=2115.0n=12
DeepSeek V4 Flashup to 285B15.0n=3
Qwen3.6up to 35B55.0n=2
Agents-A1up to 35B97.0n=1

More comparisons