"Drawing" the Mona Lisa with GPT-5.6, Claude, Gemini, and Grok

用GPT-5.6、Claude、Gemini和Grok“画”出《蒙娜丽莎》

HN 145 分 · 52 条评论 · 作者 hershyb_ · 来源 www.tryai.dev · HN 讨论

【摘要】

The article introduces the "Drawing Arena," an open-source framework where AI models attempt to draw images or illustrate text prompts using a simulated colored-pencil toolset. The study tested four vision models—GPT-5.6 Sol, Claude Fable 5, Grok 4.5, and Gemini 3.6 Flash—on two famous paintings and five descriptive prompts to evaluate their artistic capabilities and tool usage.

⋯ 继续阅读请登录会员 ⋯