Flux 3 Image Klein: open weights, 5.6B parameters

Flux 3 Image Klein is a new image model that has 5.6 billion parameters and open weights. It is based on the Transformer architecture and uses a combination of a Vision Transformer and a Transformer encoder-decoder architecture. The model is trained on the ImageNet dataset and has achieved state-of-the-art results in various image classification tasks. The weights of the model are open and can be downloaded from the Hugging Face Model Hub.