M4 Max (40-core GPU, 64 GB)
14 records across 6 models (quants: Q4, Q4_G32, Q4_K - Medium, Q5, Q5_G32, Q6, Q8).
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M4 Max (40-core GPU, 64 GB) | llama-3-1-8b-instruct | Q4_K - Medium | llamafile | 47.095953944444446 | 557.1200590555555 | – | 2.1445155647222225 | – | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | llama-3-2-1b-instruct | Q4_K - Medium | llamafile | 180.24040433333337 | 3857.499544666666 | – | 0.30436414811111107 | – | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 30B-A3B | Q4 | MLX | 92.09 | 822.59 | – | – | 2048 | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 30B-A3B | Q5 | MLX | 84.89 | 819.83 | – | – | 2048 | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 30B-A3B | Q6 | MLX | 76.74 | 817.63 | – | – | 2048 | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 30B-A3B | Q8 | MLX | 52.58 | 772.59 | – | – | 2048 | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 32B | Q4_K - Medium | Ollama | 22 | – | – | – | 128000 | – | community | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 4B | Q4_G32 | MLX | 149.07 | 2838.42 | – | – | 2048 | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 4B | Q4 | MLX | 148.08 | 2976.68 | – | – | 2048 | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 4B | Q5 | MLX | 143.2 | 2736.15 | – | – | 2048 | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 4B | Q5_G32 | MLX | 142.98 | 2754.53 | – | – | 2048 | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 4B | Q6 | MLX | 136.58 | 2735.69 | – | – | 2048 | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | Qwen 3 4B | Q8 | MLX | 111.55 | 1780.63 | – | – | 2048 | – | sourced | – | source |
| M4 Max (40-core GPU, 64 GB) | qwen-2-5-14b-instruct | Q4_K - Medium | llamafile | 25.874467866666663 | 286.74860906666663 | – | 4.498645011088889 | – | – | sourced | – | source |
All rows cite their source. Provenance: community 40018 sourced 1085