How we got this number
Each generated token reads the active part of the model from memory once: 2.5 GB. The Mac mini M4 (16GB) moves 120 GB/s, so the ceiling is 48 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 Nemotron 3.5 lightning 30B (A3B)
Other models on the Mac mini M4 (16GB)
| lfm 2.5 8B (A1B) | Fast | 78 tok/s |
| DeepSeek-R1 8B | Comfortable | 19 tok/s |
| Qwen 3 8B | Comfortable | 19 tok/s |
| Granite 4.2 8B | Comfortable | 18 tok/s |
| Granite 4.1 8B | Comfortable | 18 tok/s |
| Gemma 2 9B | Comfortable | 18 tok/s |