Skip to content

AuraFlow v0.3

An open flow-based text-to-image model

Visit official site

What AuraFlow v0.3 is

The model is described as a fully open flow-based text-to-image model trained with more compute than v0.2 and fine-tuned on an aesthetic dataset.

What you can do with it

  • Generate images at varied aspect ratios
  • Run generation with half precision and CUDA acceleration
  • Follow the provided usage examples

Application scenarios

  • A designer makes a poster with a set theme and style
  • An artist creates visual artwork with the model
  • A researcher runs academic work on image generation

Pros & cons

✓ What we like

  • Detailed usage examples are provided
  • Feedback can be shared through Discord

! What to watch out for

  • The model is still in beta
  • Generation needs width, height and step settings

FAQ

What sizes are named?

The card lists varied aspect ratios with width and height up to 1536 pixels.

Is it open?

The material describes it as a fully open model.

Last reviewed: 2026-11-15

More AI image generation tools

View all →

How we review