Vending-Bench: Testing long-term coherence in agents | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		Vending-Bench: Testing long-term coherence in agents (andonlabs.com)
		3 points by andromaton 11 months ago \| hide \| past \| favorite \| 1 comment

andromaton 11 months ago [–]

This is the same company behind Anthropic's Titanium cube vending fridge, recently on the front page. As of now, the leaderboard is #1 Claude 4 Opus #2 (distracted?) Human

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact