M3 Pro (14-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 (14-core GPU, 36 GB) | llama-3-1-8b-instruct | Q4_K - Medium | llamafile | 21.455100444444447 | 223.43918544444443 | – | 5.651565199111111 | – | – | sourced | – | source |
| M3 Pro (14-core GPU, 36 GB) | llama-3-2-1b-instruct | Q4_K - Medium | llamafile | 88.23620972222221 | 1327.8959733333334 | – | 0.9533793772777779 | – | – | sourced | – | source |
| M3 Pro (14-core GPU, 36 GB) | qwen-2-5-14b-instruct | Q4_K - Medium | llamafile | 12.10437772222222 | 119.77962366666664 | – | 10.648140152777778 | – | – | sourced | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085