How we got this number
Each generated token reads the whole model from memory once: 9 GB. The Mac mini M4 (16GB) moves 120 GB/s, so the ceiling is 13 tokens/sec. Real machines reach a fraction of that ceiling; we use 80%, calibrated against a measured Mac run. Send us your real numbers and we'll replace the estimate.
Faster machines for Qwen2.5 14B
| Mac Studio M5 Ultra (96GB) | 109 tok/s | Check price → |
| Mac Studio M5 Ultra (256GB) | 109 tok/s | Check price → |
| Mac Studio M5 Max (40-core GPU, 64GB) | 55 tok/s | Check price → |
| Mac mini M5 Pro (24GB) | 27 tok/s | Check price → |
Other models on the Mac mini M4 (16GB)
| lfm 2.5 8B (A1B) | Fast | 78 tok/s |
| Qwen2.5 3B | Fast | 51 tok/s |
| Qwen2.5 Coder 3B | Fast | 51 tok/s |
| Qwen3.5 2B | Fast | 51 tok/s |
| Qwen 3-VL 2B | Fast | 51 tok/s |
| Llama 3.2 3B | Fast | 48 tok/s |