Event ·Multi-source

Astra在浏览器使用基准测试中领先

Astra在浏览器使用基准测试v2中得分77.3%，大幅领先Opus 5的50.5%和GPT-5.6 Sol xhigh的49.1%。Astra完成了22/60项任务并获得满分，而Opus 5没有一项获得满分。测试结果显示Astra在浏览器操作任务上表现突出。

Current read

Astra在浏览器基准测试中碾压其他模型，77.3%的分数远超对手，22项任务满分

11 sources73° AI heatLast updated 2026/9/6 04:17:41

Evidence

11 sources

Primary sourceBrowser Use

2026/9/6 04:17:41View source

Related source 1向阳乔木

2026/9/4 08:57:07View source

Related source 2Decoder

2026/9/4 17:23:35View source

Related source 3Simon Willison

2026/9/4 22:04:42View source

Related source 4Perplexity

2026/9/4 22:43:47View source

Related source 5Guillermo Rauch

2026/9/5 17:11:16View source

Related source 6IT之家

2026/9/6 06:46:16View source

Related source 7AI Will

2026/9/7 03:54:08View source

Related source 8Latent Space (swyx)

2026/9/4 05:18:11View source

Related source 9orange.ai

2026/9/4 09:25:25View source

Related source 10e27

2026/9/4 09:49:04View 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