Images, documents and audio as input
Your users have screenshots, scanned receipts, 80-page PDFs and voice notes, not tidy text. Most frontier models can take all of them, and each one has a price and a few ways to go wrong.
In this lesson you'll learn to
- Send an image with a question, and estimate its token cost from its size.
- Send a PDF and explain how pages are read and billed.
- Choose between transcription and native audio input for a voice feature.
- Avoid the usual multimodal mistakes, from misread numbers to oversized inputs.
Create a free account to open this lesson
It's free. You'll get every free lesson, quizzes that save your progress, daily flashcards and streaks.