M3 Pro (18-core GPU) — LLM Benchmarks

Measured LLM inference benchmarks for M3 Pro (18-core GPU). Tokens per second across 1 models and multiple quantizations. Real runs, not estimates.

1Benchmark rows

1Models tested

30.7Fastest avg tok/s (Llama 2 7B)

0Factory-lab verified rows

Benchmark results for M3 Pro (18-core GPU)

Rows sorted by avg tok/s descending. Click source badge to see original measurement page.

Model	Quant	RAM req.	Context	Avg tok/s	Prompt tok/s	Runtime	Source
Llama 2 7B	Q4_0	3.6 GB	512	30.7 tok/s	341.7 tok/s	llama.cpp	ref

Models tested on this chip

Data

benchmarks.json — full dataset · chips.json — chip summaries · benchmarks.csv — CSV export

Data sourced from factory lab measurements and community reference runs. See all chips →