Question

A business analyst is building a lightweight app in Azure AI Foundry. Users upload a screenshot of a sales chart, and the app should answer typed questions such as 'Which quarter had the highest revenue?' based directly on what appears in the uploaded image. Which approach best fits this requirement?

Accepted Answer

Send both the image and the text question to a deployed multimodal model that can interpret visual input in the prompt

Answer

Deploy a text-only generative model and paste the chart's numeric data as text before asking the question

Answer

Use an image-generation model to recreate the chart and then answer from the new image

Answer

Use a speech-to-text model to transcribe the chart labels and answer from the transcript