Skip to content

Rig score · mac

Device readout 2026-08-03
Rig Score B tier
49%
Apple M3 Pro (18GB) · 18 GB
runs 66 of 135 models · biggest Phi-4 14B
localmodel.run/rig

Apple M3 Pro (18GB)

Runs 66 of 135 models (49%)

Strong, runs mid-size models up to ~14B.

Biggest it can run: Phi-4 14B

Grade
B
Runnable
66
Total
135
Coverage
49%

What Apple M3 Pro (18GB) can run

At 18 GB of unified memory, Apple M3 Pro (18GB) runs 66 of 135 tracked local models, 49% coverage that earns a B. The largest it handles is Phi-4 14B at 14B parameters; anything bigger needs more memory or a lower quantization.

Apple Silicon shares one 18 GB pool between CPU and GPU, so almost all of it is available to a model. MLX and llama.cpp Metal builds give the best speed here; close memory-heavy apps before loading a large model.

65 run well 1 tight fit 69 too big

Top models on Apple M3 Pro (18GB)

FAQ

What local AI rig score does Apple M3 Pro (18GB) get?

Apple M3 Pro (18GB) scores B: it runs 66 of 135 tracked local LLMs (49%). The biggest it can run is Phi-4 14B.

How big a model can Apple M3 Pro (18GB) run?

Apple M3 Pro (18GB) runs models up to Phi-4 14B (14B parameters) at Q4_K_M. Its memory is shared with the system, so leave a few GB for the OS.

Is Apple M3 Pro (18GB) good for running local LLMs?

Apple Silicon shares one 18 GB pool between CPU and GPU, so almost all of it is available to a model. MLX and llama.cpp Metal builds give the best speed here; close memory-heavy apps before loading a large model.

Want your own? Check your rig on the homepage →

Score from coverage of 135 tracked models, estimated with our memory methodology.