Google began public beta testing of its text-image AI system through its AI Test Kitchen app to gather feedback. Google announced its text-to-image AI model Imagen earlier this year, and its generation effect is comparable to OpenAI’s DALL-E 2 or Stability AI’s Stable Diffusion. OpenAI has opened DALL-E 2 to the public for use, and Stable Diffusion is even more Attract a large number of developers and users through open source. Imagen has not been made available to the public until now. Unlike DALL-E 2 and Stable Diffusion, Google imposes strict restrictions on the interaction between users and Imagen, users can not create text images as they wish, but can only add different styles on selected topics. Google product executives say this is intentional, to gather public feedback on the AI system and discover how users can push its limits.
This article is reprinted from: https://www.solidot.org/story?sid=73258
This site is for inclusion only, and the copyright belongs to the original author.