elvis · @omarsar0 · X·2026-09-04 03:55·27分钟前
AI 导读

Elvis Saravia 引用 OpenAI 官方内容评论 GPT-6 Astra:该模型在 FrontierMath Tier 4、ARC-AGI 3、TerminalBench-4.0 上达到 SOTA,并在 Terminal-Bench Science 0.1 与 HealthBench Pro 上领先。作者对表格数据表示惊讶,称其击败了几天前刚发布的竞品模型,并说 GPT-6 Astra 有些不同寻常。图中显示 ARC-AGI-3 得分 99.9%、FrontierMath Tier 4 (v2) 97.6%。

elvis@omarsar0
68AI 编辑部评分,满分 100
2026-09-04 03:55· 27分钟前
AI 导读

Elvis Saravia 引用 OpenAI 官方内容评论 GPT-6 Astra:该模型在 FrontierMath Tier 4、ARC-AGI 3、TerminalBench-4.0 上达到 SOTA,并在 Terminal-Bench Science 0.1 与 HealthBench Pro 上领先。作者对表格数据表示惊讶,称其击败了几天前刚发布的竞品模型,并说 GPT-6 Astra 有些不同寻常。图中显示 ARC-AGI-3 得分 99.9%、FrontierMath Tier 4 (v2) 97.6%。

I had to look twice at this table.

I know they are just numbers, but how the heck do you beat a competitor model released just a couple of days ago on some of the toughest benchmarks out there?

Something is different about GPT-6 Astra.

OpenAIGPT-6 Astra is state-of-the-art on FrontierMath Tier 4, ARC-AGI 3, and TerminalBench-4.0. GPT‑6 Astra is also a major advance for scientific discovery, with sta...