Four vision models (GPT-5.6 Sol, Claude Fable 5, Grok 4.5, and Gemini 3.6 Flash) were tested on drawing the Mona Lisa and Van Gogh's Starry Night, with varying degrees of success, and scored objectively using structural similarity (SSIM) metrics. The models' tool usage, cost, output, and ability to improve their work were evaluated. AI summary
Firehose
Filtered to Hacker News, tagged “deepfakes” · clear filters
Browse: People · Companies · Papers · Podcasts · Hacker News