How we got this number
Each generated token reads the whole model from memory once: 4.4 GB. The Mac mini M2 (16GB) moves 100 GB/s, so the ceiling is 23 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 starCoder 2 7B
Other models on the Mac mini M2 (16GB)
| Qwen3.5 0.8B | Fast | 80 tok/s |
| DeepSeek-R1 1.5B | Fast | 73 tok/s |
| Qwen 1.8B | Fast | 73 tok/s |
| lfm 2.5 8B (A1B) | Fast | 65 tok/s |
| Llama 3.2 1B | Fast | 62 tok/s |
| Qwen 3 1.7B | Fast | 57 tok/s |