gemma 4 12B it nvfp4 NVFP4 quantized version of google/gemma 4 12B it (12B params, unified multimodal model). Produced and maintained by vrfai. Quantization Details This model was quantized using NVIDIA ModelOpt with the following configurations: Property Value Base model google/gemma 4 12B it Quant method NVIDIA ModelOpt (NVFP4) Weight scheme 4 bit float, block size 16 Input activation 4 bit float, block size 16 Calibration dataset CNN DailyMail (512 samples, max seq len 1024) Size ~11 GB (vs ~23 GB BF16) Excluded from Quantization The following modules are kept in full precision (BF16) to preserve accuracy: lm head model.embed vision model.embed audio All self attn layers (layers 0–47) Quantization Script The recipes and scripts used to quantize this model can be found in the following repository: VinRobotics/model quantization recipes
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy