> For the complete documentation index, see [llms.txt](https://docs.avastudio.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.avastudio.com/models-and-credits/image-models.md).

# Image models

Image models create static ads, reference images, product concepts, backgrounds, and visual assets for later video generation.

Ava supports image models such as GPT Image 2, GPT Image, Nano Banana 2, Nano Banana Pro, Seedream v5 Lite, and Nano Banana. Some models are best for text-to-image. Others support image-to-image with multiple references. If your workspace shows a newer GPT Image label, follow the label in the model picker.

For a popular from-scratch image path, start a New Images project, choose Nano Banana Pro or the GPT Image family where available, write a clear composition prompt, add product or brand references if needed, generate variants, and save the strongest image to Library.

Use text-to-image when you want a new concept. Use image-to-image when product accuracy, actor resemblance, or brand style matters. For exact ad copy, use Editor Mode overlays after generation when possible.

Review quality settings where available. Higher quality may cost more or take longer, but can be worth it for hero images, product close-ups, and paid ad assets.
