Gemini 4 Carbon takes MathArena crown, costing 15 times more than GPT-6.1 Sol
Key point
In the MathArena math benchmark, Google's Gemini 4 Carbon and Anthropic's Fable 5.2 have overtaken GPT-6.1 Sol to take the top spot. However, both models cost significantly more than GPT-6.1 Sol.
In the MathArena math benchmark, Google's Gemini 4 Carbon and Anthropic's Fable 5.2 have knocked GPT-6.1 Sol from the top spot. This marks the first time in a while that no OpenAI model leads the benchmark.
Gemini 4 Carbon reportedly ignored the time limit during testing and almost never gave up on a task. However, Gemini 4 Carbon costs 15 times more than GPT-6.1 Sol, and both Gemini 4 Carbon and Fable 5.2 cost significantly more than GPT-6.1 Sol.
This item is a summary. The full, checkable version is at the original source:
Read the original at The Decoder ↗