Gemma 4 E4B IT AutoRound AWQ 4 bit This repository contains an AutoRound AWQ 4 bit quantization of google/gemma 4 E4B it . Quantization summary Method: AutoRound AWQ Bit width: 4 bit Group size: 128 Iterations: 500 Quantized block: model.language model.layers Preserved in higher precision: vision tower , audio tower , embed vision , embed audio , lm head Validation This checkpoint was smoke tested with the Transformers AWQ loader and generated the expected response to a simple text prompt. Loader note Use the Transformers AWQ loader. The working path that was validated is: Size Approximate on disk size: 9.9G Caveat This is a mixed FP/AWQ multimodal checkpoint. Runtime compatibility depends on loader support for modules to not convert in the quantization config.
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy