M2 (8-core GPU, 16 GB) — LLM Benchmarks

Measured LLM inference benchmarks for M2 (8-core GPU, 16 GB). Tokens per second across 3 models and multiple quantizations. Real runs, not estimates.

3Benchmark rows

3Models tested

35.3Fastest avg tok/s (Llama 3.2 1B Instruct)

0Factory-lab verified rows

Benchmark results for M2 (8-core GPU, 16 GB)

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 3.2 1B Instruct	Q4_K - Medium	—	—	35.3 tok/s	523.2 tok/s	—	ref
Llama 3.1 8B Instruct	Q4_K - Medium	—	—	12.9 tok/s	114.0 tok/s	—	ref
Qwen 2.5 14B Instruct	Q4_K - Medium	—	—	7.0 tok/s	60.0 tok/s	—	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 →