Yuchen Jin@Yuchenj_UW

AI 编辑部评分，满分 100

2026-09-23 12:04· 1小时前

AI 导读

今天试了 Opus 5.5。说实话没被惊艳到。 - 让它研究一个我最近学到的棘手问题。它答错了。Astra 答对了。 - 编程方面，Astra 感觉还是更好。这印证了我之前的看法：前沿 LLM 的编程能力已经停滞。现在的竞争越来越是每美元智能的比拼。

Tried Opus 5.5 today. Not blown away tbh.

• asked it to research something tricky I recently learned. It got it wrong. Astra got it right. • for coding, Astra still feels better.

This reinforces my earlier take: frontier LLM coding capability has plateaued.

The battle now is increasingly about intelligence per dollar.

来源：Yuchen Jin· x.com