Arena updates Image-to-WebDev leaderboard: GPT-6 Astra tops at 1733

Arena 更新 Image-to-WebDev 榜单：GPT-6 Astra 以 1733 分登顶

Arena added four new models to its Image-to-WebDev leaderboard. GPT-6 Astra (Max) leads at 1733, 129 points ahead of GPT-5.6 Sol (xHigh). Claude Fable 5.1 (Max) is third at 1710, Muse Spark 1.3 (Max) fourth at 1645, and GLM-5.3-Flash tenth at 1588. The post doesn't disclose evaluation tasks or sample size, so I'd take the gaps with a grain of salt.

Why it matters: GPT-6 Astra tops the Image-to-WebDev leaderboard on its first appearance with a meaningful margin — newsworthy. But the post only gives scores and rankings, with no detail on methodology, task difficulty, or model differences. H and K both hit, R is absent — lands right at the...