experimental_generateImage token usage

[Morriz]

Description

Currently experimental_generateImage does not return usage date. Is that coming soon? I hope so...

AI SDK Version

• ai: 5.0.27
• @ai-sdk/openai: 2.0.22
• @ai-sdk/xai: 2.0.13

Code of Conduct

• I agree to follow this project's Code of Conduct

[gr2m]

Would you like to have a look at it on why token usage is not exposed yet? Might be a simple fix. A good starting point would be a pull request adding or updating an example (like examples/ai-core/build/generate-image/openai.js) to showcase what you would like to see working, and we can take it from there?

[kingsleytorlowei]

I'm looking into this. Something like this would be the expected usage right?

import { openai } from '@ai-sdk/openai';
import { experimental_generateImage as generateImage } from 'ai';
import 'dotenv/config';

async function main() {
  const prompt = 'Santa Claus driving a Cadillac';
  const result = await generateImage({
    model: openai.image('dall-e-3'),
    prompt,
  });

  console.log(result.usage);

}

main().catch(console.error);

[jonaslalin]

I think it would make sense to update the v3 spec for packages/provider/src/image-model/v2/image-model-v2.ts to also have something like packages/provider/src/language-model/v2/language-model-v2-usage.ts. We also need usage tracking for image generation.