Ideogram 4.5 文生图：Knowledge、Physics 与复杂构图最接近前沿
Where Ideogram 4.5 is strongest in Text to Image: closest to the category frontier in Knowledge, Physics and Complex Compositions
Artificial Analysis 的 Text to Image 评测显示，Ideogram 4.5 在 Knowledge 能力上最接近类别前沿，其次是 Physics 和 Complex Compositions。相比 Ideogram 4.0（Quality），它在 9 项能力中的 5 项缩小了与前沿的差距，其中 Text Rendering 和 Complex Compositions 提升最大。该评测体系涵盖 9 项能力和 10 类用例，各设独立排行榜。
正文
当前显示原文，中文全文译文尚未就绪。
Where Ideogram 4.5 is strongest in Text to Image: closest to the category frontier in Knowledge, Physics and Complex Compositions
Our Text to Image taxonomy measures 9 capabilities and 10 use cases, each with its own leaderboard. Capabilities are the individual skills that go into making an image.
Ideogram 4.5 is closest to the category frontier in Knowledge, followed by Physics and Complex Compositions. Against Ideogram 4.0 (Quality) it closes the gap to the frontier on five of the nine capabilities, with the largest gains in Text Rendering and Complex Compositions.
➤ Knowledge covers real landmarks, species, and domain facts across science and common sense.
➤ Physics covers gravity, support, collision, and thermal and state change.
➤ Complex Compositions covers precise counting, spatial relationships, subject interactions, and attribute binding.