Event ·Multi-source

Canva 发布自研图像模型，DesignArena 排名仅次于 OpenAI

Canva 推出首个自研图像生成模型 Canva Craft，在 DesignArena 图像生成榜单上位列第二，仅次于 OpenAI。该模型采用两阶段架构：先由规划模型读取提示词，写出画面中各元素的位置、大小、强调点和需显示的文字；再由渲染模型按计划生成最高 2K 分辨率、支持 30 多种宽高比的图像。文字拼写和排版在渲染前就已确定，修改某个元素时只有该元素变化，不会影响旁边的其他内容。

Current read

做设计工具的 Canva 也自己训模型了，图像生成榜排名第二，先把排版和文字规划好再渲染，改图只动改的那块。

11 sources71° AI heatLast updated 2026/10/22 20:34:15

Evidence

11 sources

Primary sourcerohanpaul_ai

2026/10/22 20:34:15View source

Related source 1IT之家

2026/10/20 23:45:48View source

Related source 2Liputan6 Tekno

2026/10/21 04:11:49View source

Related source 3Tanishq Abraham (论文推介)

2026/10/21 04:28:16View source

Related source 4Pandaily

2026/10/21 06:49:49View source

Related source 5kimmonismus

2026/10/21 06:53:40View source

Related source 6The Information

2026/10/21 14:00:28View source

Related source 7Andrew Curran

2026/10/21 14:25:37View source

Related source 8Artificial Analysis

2026/10/21 15:53:56View source

Related source 9Genk

2026/10/21 16:38:00View source

Related source 10TestingCatalog

2026/10/21 18:57:16View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story