The new version of the AI image generation model from Ideogram — 2.1, is currently in the early testing stage. The main improvement in version 2.1 is its ability to generate consistent text within images, which remains a challenge for many AI models. TestingCatalog shared examples where all four test results showed high text consistency, which could be a significant step forward in this field.
The context provided by lamarena demonstrates the effectiveness of Ideogram 2.0 on their leaderboard. The service took first place in the “User Prompts Only” category and third in the “Pre-generated Prompts” category. This indicates that the model performs well with creative or specific prompts but faces more competition when working with standardized prompts. The new version, 2.1, aims to improve these capabilities, particularly focusing on text consistency within generated images.
Although there has not yet been an official announcement about Ideogram 2.1, its development aligns with general trends in the generative AI field. Improvements in prompt adherence and specific use cases, such as text generation in images, are becoming key areas of innovation. This feature could be especially valuable for industries like advertising, design, and content creation, where accurate text rendering in visuals is crucial.
Ideogram’s progress also reflects the growing competition in the generative AI space. According to the lamarena.ai leaderboard, other models such as Recraft V3 are also showing strong results, indicating a competitive environment where constant improvements are necessary to maintain leadership.
