A compact, quantized variant of Google's Gemma 4 family, running at 4-bit weights with 16-bit activations (W4A16). The quantization makes it notably leaner in memory footprint compared to its full-precision counterparts, with the typical trade-off of slight accuracy degradation in exchange for faster inference and lower hardware requirements. It handles text-in, text-out tasks and is suited for deployment in resource-constrained environments.