vrfai/gemma-4-12B-it-fp8

FP8 quantized version of google/gemma-4-12B-it (12B params, unified multimodal model). Produced and maintained by vrfai. This model was quantized using NVIDIA ModelOpt with the following configurations: The following modules are kept in full precision (BF16) to preserve accuracy: — lmhead — model.embedvision — model.embedaudio — All selfattn layers (layers 0–47) The recipes and scripts used to quantize this model can be found in the following repository: — VinRobotics/model-quantization-recipes

Модальности:
Генерация текста

Области применения:
Диалог / чат

Задача: Генерация текста
Автор: vrfai
Теги: gemma4_unified, image-text-to-text, fp8, quantized, gemma4, vrfai, conversational, endpoints_compatible
Лайков: 3  |  Загрузок: 287

Открыть на HuggingFace →

Описание основано на материалах HuggingFace. Перевод выполнен автоматически.