llamaperf

M4 Max 128GB vs RTX 4090

For running local LLMs · 22 reports across 6 models

Side A
M4 Max 128GB
Vendor
apple
VRAM
128GB
Memory
Unified
Side B
RTX 4090
Vendor
nvidia
VRAM
24GB
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 →

ModelM4 Max 128GBRTX 4090
Qwen3.8up to 2400B15.2n=2134.0n=7
DeepSeek V4 Flashup to 285B180.0n=6
Qwen3.6up to 35B72.5n=180.0n=2
Gemma 4up to 31B149.6n=2
GLM-5.3n=1
Museup to 30B25.0n=1

More comparisons