1. What This Means in the Exam
Using Azure AI Foundry, you should recognise how to:
Send both text + image in a prompt
Use a multimodal model to understand the visual content
Generate a text response based on the image
👉 Think:
“Show the AI an image and ask it a question about it”
🧩 2. What is a Multimodal Model?
A multimodal model can process multiple input types at once, such as:
Text
Images
(Sometimes audio)
👉 In this scenario:
Input = Image + text prompt
Output = Text (interpretation/explanation)
🖼️ 3. Example of Visual Input in Prompts
Example prompts:
đź“Ś Scenario 1
“What items are listed in this receipt?”
✔️ Model interprets:
Text within image
Returns structured answer
đź“Ś Scenario 2
“Describe what is happening in this image”
✔️ Model:
Analyses objects and context
Generates description
đź“Ś Scenario 3
“Is there any defect in this product image?”
✔️ Model:
Detects anomaly/issue
🔄 4. How It Works (Exam-Friendly Flow)
Image + Text Prompt
↓
Multimodal Model (in Foundry)
↓
Understands visual + textual context
↓
Generates response (text)
⚙️ 5. How It’s Implemented in Foundry
Using Azure AI Foundry:
Deploy a multimodal model
Send:
Image (file or URL)
Text prompt
Model processes both inputs
Returns a response
⚖️ 6. Key Exam Distinctions
Capability Multimodal Interpretation Computer Vision
Input Image + text Image only
Output Natural language response Structured data (labels, OCR, etc.)
Use case Ask questions about image Analyse image features
⚠️ 7. Common Exam Traps
❌ Thinking this is just OCR
✔️ It’s broader—understanding context, not just text
❌ Confusing with image generation
✔️ No image is created—only interpreted
❌ Ignoring text prompt
✔️ Prompt guides what the model looks for
🧠8. Simple Memory Trick
“Show + Ask → Answer”
Show → image
Ask → text prompt
Answer → model response
🎯 9. Example Exam Scenario
Scenario:
“A user uploads a photo of a chart and asks for a summary of trends”
✔️ Use:
Multimodal model
Interpret visual input
Generate explanation
âś… Summary
To interpret visual input in prompts using a deployed multimodal model:
Use Azure AI Foundry
Combine:
Image input
Text prompt
Model:
Understands visual + textual context
Generates response