How we got this number
Measured on a real Mac mini M2 (16GB) with Ollama (VsMacs run, Ollama 0.34.4, Sep 2026).
Faster machines for Qwen 3 8B
Other models on the Mac mini M2 (16GB)
| 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 |
| Granite 3.1 moe 1B | Fast | 57 tok/s |