M3 Pro (14-core GPU, 18 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 (14-core GPU, 18 GB) | llama-3-1-8b-instruct | Q4_K - Medium | llamafile | 19.10761944444444 | 199.51528166666668 | – | 6.427559875000002 | – | – | sourced | – | source |
| M3 Pro (14-core GPU, 18 GB) | llama-3-2-1b-instruct | Q4_K - Medium | llamafile | 88.10604988888889 | 1343.9570948888888 | – | 0.9594814676666664 | – | – | sourced | – | source |
| M3 Pro (14-core GPU, 18 GB) | qwen-2-5-14b-instruct | Q4_K - Medium | llamafile | 11.886003166666669 | 117.03684152777775 | – | 10.918680665472221 | – | – | sourced | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085