Microsoft · AI-901
Vision and image generation apps
AI-901 · Lesson 9 of 10 · 55:42
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Prepare for the AI-901 exam (Microsoft Azure AI Fundamentals). In this lesson: sending images to a vision-capable model with the Responses API (input_text and input_image parts, a public URL or a base64 data URL, the detail setting), generating images with images.generate and a GPT-image deployment (size, quality, n, base64 output), the image_generation tool of the Responses API, and a small Python vision app built on a multimodal model. Includes three original practice questions.
Preview the topics
- Customers upload photos of damaged parcels to a backend with no public storage. It must ask a vision model whether each box is crushed. How should it send each image?
- After calling images.generate with a GPT-image deployment, how do you save the picture?
- An app sends thousands of small thumbnails to a vision model, only to label each one photo or illustration. Which setting cuts tokens and latency?