gemma-4-e2b-qat

Gemma 4 E2B QAT

Gemma 4 E2B QAT is the Quantization-Aware Training version of Gemma 4 E2B. It aims to keep quality close to bfloat16 while using much less memory to load the model.

Gemma 4 is an open multimodal model family from Google DeepMind. It supports text and image input, text output, reasoning, long context, system prompts, and native tool use.

Gemma 4 E2B is one of the lightest Gemma 4 models. It is designed for efficient local use, and the QAT build makes it easier to run on tighter memory budgets.