Chubby♨️ · @kimmonismus · X·2026-09-09 05:32·38分钟前
AI 导读

OpenAI 自 8 月 28 日起训练的新内部模型,仅用一周便在数学等基准测试中全面超越其最强模型 GPT-6-Astra-xhigh。该模型训练仍在进行,性能持续提升,作者认为外界尚未充分理解其意义。

Chubby♨️@kimmonismus
34AI 编辑部评分,满分 100
2026-09-09 05:32· 38分钟前
AI 导读

OpenAI 自 8 月 28 日起训练的新内部模型,仅用一周便在数学等基准测试中全面超越其最强模型 GPT-6-Astra-xhigh。该模型训练仍在进行,性能持续提升,作者认为外界尚未充分理解其意义。

To reiterate: training on OpenAI's new model began on August 28th.

It took just one week, seven days, to completely outperform OpenAI's best model, GPT-6-Astra-xhigh.

One week. Thats all it takes nowdays to outperform sota. I think people still dont fully grasp what OpenAI actually wrote in their blogpost.

Chubby♨️"Since August 28 we have been training a new internal model that has exhibited unprecedented performance in our benchmarks, including mathematics. This model’s ...

来源:Chubby♨️· x.com