Event ·Multi-source

Grok 4.7 浏览器长任务基准 39.9 分，仍落后 DeepSeek

browser_use 官方发布 Long Horizon Browser Use Benchmark v2 榜单，纳入刚发布的 Grok 4.7。Grok 4.7 得分 39.9，高于 Grok 4.6 的 31.2，但不到 GPT-6 Astra 80.6 分的一半。DeepSeek V4.1 Flash 以 47.9 分排第二，领先 Grok 4.7 约 8 分。

Current read

Grok 4.7 浏览器任务 39.9 分，比上代涨 8.7，但只有 GPT-6 Astra 一半。

11 sources22° AI heatLast updated 2026/9/21 20:53:59

Evidence

11 sources

Primary sourceBrowser Use

2026/9/21 20:53:59View source

Related source 1shao__meng

2026/9/23 01:44:23View source

Related source 2宝玉

2026/9/21 18:12:28View source

Related source 3Blognone

2026/9/22 01:30:57View source

Related source 4IT之家

2026/9/20 05:46:36View source

Related source 5AI Will

2026/9/20 08:04:15View source

Related source 6arXiv: DeepSeek

2026/9/21 06:39:47View source

Related source 7Justine Moore

2026/9/21 18:00:21View source

Related source 8eric zakariasson

2026/9/21 22:01:00View source

Related source 9Simon Willison’s Weblog

2026/9/22 23:46:41View source

Related source 10歸藏(guizang.ai)

2026/9/21 03:06:58View 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