Summary

  • AI start-up Reve AI has launched Reve Image, its first generation model for the creation of images from text descriptions.
  • As well as generating images, the new model allows users to modify existing images using simple language commands, and is the only one on the market that supports the uploading of reference images as a guide.
  • Third-party testing service Artificial Analysis currently has Reve Image topping its benchmarking charts for image generation quality.
  • While the model is not yet available via API or open source, users are able to access it for free and provide feedback at preview.reve.art.
  • Future developments are expected to include API access, long-term pricing plans, custom model training, animation controls and integration with creative software.
  • The company’s vision is to build AI models that understand creative intent rather than just generating visually plausible outputs.

By Carl Franzen

Original Article