Gen-4.5 generates videos from text descriptions or static images with emphasis on realistic motion quality and visual consistency. The model excels at interpreting detailed prompts and maintaining coherent object behavior throughout sequences. Suitable for creative projects, visual effects planning, and content generation workflows.