Can I add images when using VLMs？

I wonder if I can add images when using gpt4o for html-coding. Currently I cannot find a way to input an image using llm functions in datadreamer

background：I would like to prompt gpt4o to mimic the layout of the provided image