marco (multimodal attentive residual co-transformer)

A multimodal language model capable of understanding and generating language based on text and image inputs.