M3 Max (40-core GPU, 64 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 Max (40-core GPU, 64 GB) | llama-3-1-8b-instruct | Q4_K - Medium | llamafile | 25.448832000000003 | 377.04858177777777 | – | 3.3699854838888887 | – | – | sourced | – | source |
| M3 Max (40-core GPU, 64 GB) | llama-3-2-1b-instruct | Q4_K - Medium | llamafile | 106.97443855555557 | 2521.617585666667 | – | 0.520643525388889 | – | – | sourced | – | source |
| M3 Max (40-core GPU, 64 GB) | qwen-2-5-14b-instruct | Q4_K - Medium | llamafile | 13.801572 | 200.82679227777783 | – | 6.657162884277778 | – | – | sourced | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085