# Mixed precision training

Kanana Nano 2.1B Instruct is a 2.1 billion parameter instruction-finetuned language model developed by Kakao, supporting English and Korean text generation tasks.

This is an audio processing model fine-tuned based on facebook/wav2vec2-base, mainly used for speech-related tasks

A fill-mask model fine-tuned based on xlm-roberta-base for predicting masked text segments

Vision Transformer model fine-tuned from google/vit-base-patch16-224, suitable for image classification tasks

This model is a fine-tuned Vision Transformer based on google/vit-base-patch16-224-in21k on an unknown dataset, primarily used for image classification tasks.