Models76°

Ethan Mollick 称 GPT-6.2 Astra 难驾驭但大项目表现出色

I’ve been using GPT-6.2 Astra prior to launch &amp…

Why we picked it

Ethan Mollick 提前用了 GPT-6.2 Astra,说它比别的模型难调教,但让他用 Blender 和 Unity 从零做太空电梯场景,完成得挺好。如果你关心新模型实际好不好用、该怎么测,这篇一手体验比跑分更有参考价值。

Ethan Mollick 在 X 上分享了他在 GPT-6.2 Astra 发布前的实际使用体验。他的评价并不一致:一方面,Astra 比其他模型更难引导和协作,日常使用中感觉“不均匀且怪异”;另一方面,在给他一个高难度项目时——从零用 Blender 和 Unity 做一个太空电梯乘坐场景——它能交付不错的结果。

Body

这条体验指出了 Astra 的一个矛盾点:常规任务下服从性、可控性偏弱,但在范围大、需要长链路执行的野心项目上反而能撑住。也就是说,问题可能不在能力上限,而在它与用户指令的配合方式。

对开发者和重度用户来说,这个观察有参考价值:评估这类模型不能只看简单问答场景,需要用真实的多步骤创作项目来测试,才能发现它在常规评测里暴露不出的优缺点。

Original · Ethan Mollick

I’ve been using GPT-6.2 Astra prior to launch &amp…

I’ve been using GPT-6.2 Astra prior to launch & I found it very uneven & weird. On one hand, it was harder to steer & work with than other models. On the other, when given an ambitious project (make a space elevator ride from scratch with Blender & Unity) it could deliver well https://t.co/gc10YRzIkw

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Models" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include GPT-6.2, Astra, OpenAI, multimodal.

Source attribution: curated from Ethan Mollick (@emollick).