Skip to main content

Home > Resources > Generate Better AI Presenters from Text

Product Updates
25 January 2026

Generate Better AI Presenters from Text

We’ve updated the image generation engine used in  D-ID Studio’s photo-based avatar creation. The model is tailored to produce characters that work in avatar videos, so you can design your presenter and animate it all in one place.

What this means for you

  • Same workflow, better outputs: the UI and creation flow are unchanged, but the update means improved image quality and more consistent results.

  • More reliable creation: the update supports more stable generation so you can get to a usable presenter image for your video or visual agent faster, with fewer retries.

Screenshot of an avatar generator showing a prompt for a female superhero with a shaved head and blue eyes, and the generated avatar image below with those features.

How works:

  1. Go to Avatars

  2. Click Generate from text

  3. Write your prompt (describe the person clearly)

  4. Click Generate

  5. Review the result

  6. Click Add avatar to use it, or Generate again to iterate