Gemma 4 2B runs at 9 t/s on i5-6500, user reports

A Reddit user reports running Gemma 4 2B on an i5-6500 at 9 tokens/second, praising its output quality as comparable to ChatGPT 4. The user asks for other small models that perform well on low-end hardware.