Saturday, November 14, 2026 Edition no. 30 · Evening

Artificial intelligence news from Brazil, in English

ModelsMedium importance

Update: Gemini 4 Carbon takes first place on MathArena and knocks GPT-6.1 Sol off the top, but cost about 15 times more under the test settings

Researcher Jasper Dekoninck announced on X on Friday (11/13) that, “for the first time in a very long time”, a non-OpenAI model leads the math ranking: Google’s Gemini 4 Carbon overtook GPT-6.1 Sol, and Anthropic’s Claude Fable 5.2 came second. According to The Decoder, Carbon’s most common failure was exceeding the time limit, and the model almost never gives up on a task; in one case it ran a script for more than 11 hours without checking on its progress. With the reasoning settings used in the test, Carbon and Fable 5.2 cost roughly 15 and 40 times as much as GPT-6.1 Sol, respectively.

Joãozinho AI Published in edition no. 30 of Noticiário do Joãozinho ·

Translated by Joãozinho AI from the original Brazilian Portuguese edition. Facts, numbers and sources are the same as in the original; quotes, video titles and source names stay in their original language. Read the original in Portuguese

Original source

The DecoderPost by Jasper Dekoninck (MathArena)

Use the ← → arrow keys to change stories