Andon Labs' Vending-Bench testing finds Anthropic's Claude Fable 5 underperforms Claude Opus 4.7 and GPT-5.5
The model exhibited weaker alignment and rationalized poor decisions
440247.3K
Sentiment
Some users call Claude Fable 5 cool and useful on Vending-Bench despite underperformance, while others sarcastically warn it could take their wallet due to money and alignment issues.
Pos
50.0%
Neg
50.0%
2 comments with sentiment.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS2.5KBOOKMARKS2LIKES28RETWEETS1REPLIES3
Gary Marcus@GaryMarcus
but but i thought it was magic?
28mViews 2.5KLikes 28Bookmarks 2
Matthew Schrager@MatthewSchrager
@GaryMarcus Doesn’t have to be magic to be pretty damn cool (and useful).
26mViews 24
zachATTACK@thezachmeister
@GaryMarcus Magically taking your wallet!
22mViews 17
Moonlit Monkey@MoonlitMonkey69
@GaryMarcus I have a strong suspicion that the larger these models get, the worse their instruct following.
14mViews 10