# QA: Multimodal demo — LlamaIndex

## Prerequisites

- Demo is deployed and accessible
- Agent backend configured with a vision-capable model (gpt-4o)

## Test Steps

- [ ] Navigate to the multimodal demo page
- [ ] Upload an image and ask a question about it
- [ ] Verify the agent answers based on the image content
- [ ] Upload a PDF and ask a question about its contents