Which Amazon Bedrock foundation model can be fine-tuned to handle text, images, and video understanding?

Amazon Nova Pro is the correct answer because it is Amazon Bedrock's most advanced multimodal foundation model, designed to handle text, images, and video understanding, and supports fine-tuning for specific tasks across these modalities. Amazon Titan Multimodal Embeddings G1 focuses on generating embeddings for multimodal search and recommendations, not general understanding and fine-tuning across all three data types. Amazon Titan Text G1 - Express is a text-only model optimized for speed and cost-effectiveness, not multimodal capabilities. Amazon Nova Micro is a fictional model; no such model exists within Amazon Bedrock.