Claude reportedly built five of seven functional shops in Stripe's T-shirt benchmark
A post sharing Breaking Change's debut article says all participating models chose to integrate Stripe, and shops built by Fable and Opus were the two finalists selected for production.
TLDR
Breaking Change's author introduces the new Substack's first article by asking whether models can build fully functioning T-shirt shops. Their answer: “Sort of.” A post sharing the article describes Stripe's UnseriousT-ShirtShopBench as asking models to build and run online stores. It says all chose to integrate Stripe, five of the seven functional shops were built by Claude, and the two finalists chosen for production were built by Fable and Opus.
Combined views
76.1K
4 Sources, first seen ago
