Record Final Object after Streaming Object

Learn how to record the final object after streaming an object using the AI TOOLKIT and Node

2 min readnodestreamingstructured dataView source

When you're streaming structured data, you may want to record the final object for logging or other purposes.

Run it locally

$ npm install ai

`onFinish` Callback

You can use the onFinish callback to record the final object.

It is called when the stream is finished.

The object parameter contains the final object, or undefined if the type validation fails.

There is also an error parameter that contains error when e.g. the object does not match the schema.

import { streamObject } from 'ai-toolkit';

import { z } from 'zod';

const result = streamObject({

model: 'openai/gpt-4.1',

schema: z.object({

recipe: z.object({

name: z.string(),

ingredients: z.array(z.string()),

steps: z.array(z.string()),

}),

}),

prompt: 'Generate a lasagna recipe.',

onFinish({ object, error }) {

// handle type validation failure (when the object does not match the schema):

if (object === undefined) {

console.error('Error:', error);

return;

}

console.log('Final object:', JSON.stringify(object, null, 2));

},

});

index.ts
ts
  1. 1import { streamObject } from 'ai-toolkit';
  2. 2import { z } from 'zod';
  3. 3const result = streamObject({
  4. 4 model: 'openai/gpt-4.1',
  5. 5 schema: z.object({
  6. 6 recipe: z.object({
  7. 7 name: z.string(),
  8. 8 ingredients: z.array(z.string()),
  9. 9 steps: z.array(z.string()),
  10. 10 }),
  11. 11 }),
  12. 12 prompt: 'Generate a lasagna recipe.',
  13. 13 onFinish({ object, error }) {
  14. 14 // handle type validation failure (when the object does not match the schema):
  15. 15 if (object === undefined) {
  16. 16 console.error('Error:', error);
  17. 17 return;
  18. 18 }
  19. 19 console.log('Final object:', JSON.stringify(object, null, 2));
  20. 20 },
  21. 21});

`object` Promise

The `streamObject` result contains an object promise that resolves to the final object.

The object is fully typed. When the type validation according to the schema fails, the promise will be rejected with a TypeValidationError.

import { streamObject } from 'ai-toolkit';

import { z } from 'zod';

const result = streamObject({

model: 'openai/gpt-4.1',

schema: z.object({

recipe: z.object({

name: z.string(),

ingredients: z.array(z.string()),

steps: z.array(z.string()),

}),

}),

prompt: 'Generate a lasagna recipe.',

});

result.object

.then(({ recipe }) => {

// do something with the fully typed, final object:

console.log('Recipe:', JSON.stringify(recipe, null, 2));

})

.catch(error => {

// handle type validation failure

// (when the object does not match the schema):

console.error(error);

});

// note: the stream needs to be consumed because of backpressure

for await (const partialObject of result.partialObjectStream) {

}

index.ts
ts
  1. 1import { streamObject } from 'ai-toolkit';
  2. 2import { z } from 'zod';
  3. 3const result = streamObject({
  4. 4 model: 'openai/gpt-4.1',
  5. 5 schema: z.object({
  6. 6 recipe: z.object({
  7. 7 name: z.string(),
  8. 8 ingredients: z.array(z.string()),
  9. 9 steps: z.array(z.string()),
  10. 10 }),
  11. 11 }),
  12. 12 prompt: 'Generate a lasagna recipe.',
  13. 13});
  14. 14result.object
  15. 15 .then(({ recipe }) => {
  16. 16 // do something with the fully typed, final object:
  17. 17 console.log('Recipe:', JSON.stringify(recipe, null, 2));
  18. 18 })
  19. 19 .catch(error => {
  20. 20 // handle type validation failure
  21. 21 // (when the object does not match the schema):
  22. 22 console.error(error);
  23. 23 });
  24. 24// note: the stream needs to be consumed because of backpressure
  25. 25for await (const partialObject of result.partialObjectStream) {
  26. 26}