M3 Pro (18-core GPU, 36 GB)
3 records across 3 models (quants: Q4_K - Medium).
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M3 Pro (18-core GPU, 36 GB) | llama-3-1-8b-instruct | Q4_K - Medium | llamafile | 22.06217488888889 | 283.8059724444445 | – | 4.460495907444445 | – | – | sourced | – | source |
| M3 Pro (18-core GPU, 36 GB) | llama-3-2-1b-instruct | Q4_K - Medium | llamafile | 89.75982788888888 | 1586.168225333333 | – | 0.8094999791666667 | – | – | sourced | – | source |
| M3 Pro (18-core GPU, 36 GB) | qwen-2-5-14b-instruct | Q4_K - Medium | llamafile | 11.988618277777778 | 147.43616580555556 | – | 8.729495730361112 | – | – | sourced | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085