Learn how to stream structured data with an image prompt using the AI SDK and Node
Some language models that support vision capabilities accept images as part of the prompt. Here are some of the different formats you can use to include images as input.
import { streamObject } from 'ai';
import dotenv from 'dotenv';
import { z } from 'zod';
dotenv.config();
async function main() {
const { partialObjectStream } = streamObject({
model: 'openai/gpt-4.1',
maxOutputTokens: 512,
schema: z.object({
stamps: z.array(
z.object({
country: z.string(),
date: z.string(),
}),
),
}),
messages: [
{
role: 'user',
content: [
{
type: 'text',
text: 'list all the stamps in these passport pages?',
},
{
type: 'image',
image: new URL(
'https://upload.wikimedia.org/wikipedia/commons/thumb/c/c5/WW2_Spanish_official_passport.jpg/1498px-WW2_Spanish_official_passport.jpg',
),
},
],
},
],
});
for await (const partialObject of partialObjectStream) {
console.clear();
console.log(partialObject);
}
}
main();import { streamObject } from 'ai';
import dotenv from 'dotenv';
import { z } from 'zod';
import fs from 'fs';
dotenv.config();
async function main() {
const { partialObjectStream } = streamObject({
model: 'openai/gpt-4.1',
maxOutputTokens: 512,
schema: z.object({
stamps: z.array(
z.object({
country: z.string(),
date: z.string(),
}),
),
}),
messages: [
{
role: 'user',
content: [
{
type: 'text',
text: 'list all the stamps in these passport pages?',
},
{
type: 'image',
image: fs.readFileSync('./data/passport.png', {
encoding: 'base64',
}),
},
],
},
],
});
for await (const partialObject of partialObjectStream) {
console.clear();
console.log(partialObject);
}
}
main();