aifollow.news
返回 Tibo
Tibo· @thsottiaux · X· · 原发布时间 AI 评分13

OpenAI 模型引发不同评价,发言者称将逐步改进不足

AI 导读

一名发言者称,这些模型能力很强,但仍有一些待改进之处;引述的另一位用户则称,Astra 6.2 在其自行运行的研究评测中,以低推理投入达到该评测的上限。材料没有给出评测方法或独立验证结果。

原帖引用 · 原文

@thsottiaux

@JoshPurtell Beauty is in the eye of the beholder, but clearly the models are polarizing. A bit like we had in the early codex days. These models are clearly extremely capable but have a couple of rough edges that we will sand down over time.

回复Josh@JoshPurtell
Days like this I genuinely feel like I live in a different universe from the rest of the world. Saw ubiquitous 6.2 hate / it's over for OpenAI yada yada ... ran the highest-signal autoresearch bench and 6.2 sol/astr are *absolute monsters*. Astra 6.2 saturates it at low effort https://t.co/YFhXQbC9dc
在 X 查看这条帖子 ↗
相关上下文Josh@JoshPurtell
@_arohan_ Astra 6 was better than Fable 5.2 and Opus 5.5 on my highest signal OOD autoresearch eval. Haven't tested Astra 6.2 but assume it's modestly better. I can't speak to internal models but do you really prefer their public models for research that much? https://t.co/nuYWwNILyj
在 X 查看这条帖子 ↗

来源:Tibo · x.com

观点
发现内容有误?提交纠错