llamaperf

M2 Max 96GB vs RTX 5070 Ti

For running local LLMs · 19 reports across 4 models

Side A
M2 Max 96GB
Vendor
apple
VRAM
96GB
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 →

ModelM2 Max 96GBRTX 5070 Ti
Qwen3.8up to 2400B22.0n=2115.0n=10
Qwen3.6up to 35B28.0n=355.0n=2
Agents-A1up to 35B97.0n=1
Qwen3up to 235B43.0n=1

More comparisons