Hacker News user asks what was the last task where only a frontier model could do it, sparking discussion on open vs closed models

A Hacker News Ask post asks users to share concrete tasks where frontier models like Opus, Fable, or GPT succeeded while open models failed, prompting a discussion on the capabilities of open-weight models.

• Ask HN: What was the last task where only a frontier model could do it?