AI article

My benchmark scored GPT-5.4 mini 1.00 by silently skipping the 3 questions it failed

Community description: This is a submission for the Kaggle Benchmarking Challenge GPT-5.4 mini got a perfect 1.00 on my...

Dev.to | Oct 9, 2026 | Dmitriy Topka

Automated excerpt

Every model failed 30-70% of the installment questions. Gemma 4: "На жаль, розстрочка на 10 місяців недоступна. Ми пропонуємо «Оплату частинами» від monobank від 2 до 3 платежів... " ("Sorry, 10 months isn't available. Claude Haiku 4. 5 apologized in perfect Russian for not speaking Russian: "Прошу прощения, я консультант украинского магазина и отвечаю на украинском языке" ("Sorry, I work for a Ukrainian shop and answer in Ukrainian", from an earlier run), then answered in Ukrainian.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.

Read the original article

Read next

OpenAI coverage

AI briefing: recent picks

More stories to explore