2 ms·

Model used: Llama 3.2 1B (small). Quality hould be similar with running Ollama app with same small model.