M1 Max (32-core GPU) — LLM Benchmarks

Measured LLM inference benchmarks for M1 Max (32-core GPU) across all RAM configurations (32 GB, 64 GB). 6 benchmark rows across 3 models. Compare how RAM affects throughput. Real runs, not estimates.

6Benchmark rows

3Models tested

2RAM configurations

125.8Fastest avg tok/s

RAM configurations

Each configuration differs only in unified memory. More RAM = larger models fit. Throughput is similar across RAM tiers at the same model size.

32 GBM1 Max (32-core GPU, 32 GB)3 rows · 3 models125.8 tok/s peak 64 GBM1 Max (32-core GPU, 64 GB)3 rows · 3 models120.7 tok/s peak

All benchmark rows — M1 Max (32-core GPU)

Sorted by avg tok/s descending. Click source badge to see original measurement.

Chip (RAM)	Model	Quant	RAM req.	Avg tok/s	Prompt tok/s	Runtime	Source
M1 Max (32-core GPU, 32 GB)	Llama 3.2 1B Instruct	Q4_K - Medium	—	125.8 tok/s	2025.2 tok/s	—	ref
M1 Max (32-core GPU, 64 GB)	Llama 3.2 1B Instruct	Q4_K - Medium	—	120.7 tok/s	1972.1 tok/s	—	ref
M1 Max (32-core GPU, 64 GB)	Llama 3.1 8B Instruct	Q4_K - Medium	—	37.8 tok/s	380.7 tok/s	—	ref
M1 Max (32-core GPU, 32 GB)	Llama 3.1 8B Instruct	Q4_K - Medium	—	35.4 tok/s	359.9 tok/s	—	ref
M1 Max (32-core GPU, 32 GB)	Qwen 2.5 14B Instruct	Q4_K - Medium	—	20.1 tok/s	195.8 tok/s	—	ref
M1 Max (32-core GPU, 64 GB)	Qwen 2.5 14B Instruct	Q4_K - Medium	—	19.0 tok/s	185.9 tok/s	—	ref

Models tested on M1 Max (32-core GPU)

Llama 3.1 8B Instruct Llama 3.2 1B Instruct Qwen 2.5 14B Instruct

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 →