FlashNEWS BRIEFING

Gemini 4 Carbon Tops MathArena Benchmark but Costs Far More Than GPT-6.1 Sol

1 min read

88Z AI, 15 November — Google's Gemini 4 Carbon and Anthropic's Fable 5.2 have surpassed OpenAI's GPT-6.1 Sol in the MathArena AI math benchmark, ending its long-standing lead. Despite their superior performance, both new models come with significantly higher costs, with Gemini 4 Carbon priced at 15 times that of GPT-6.1 Sol. Tests indicate that Gemini 4 Carbon ignores time limits and rarely gives up on tasks.

At a glance

  • Gemini 4 Carbon and Fable 5.2 surpass GPT-6.1 Sol in MathArena
  • Gemini 4 Carbon costs 15x more than GPT-6.1 Sol and ignores time limits in tests
Illustration of Gemini 4 Carbon performance in MathArena benchmark
Illustration of Gemini 4 Carbon performance in MathArena benchmark

Benchmark Leadership Shift

Google's Gemini 4 Carbon and Anthropic's Fable 5.2 have defeated OpenAI's GPT-6.1 Sol in the MathArena AI math benchmark. This marks the first time in a while that no OpenAI model holds the top spot, signaling a significant shift in the competitive landscape of mathematical reasoning.

Performance vs. Cost Trade-off

While the new models achieve breakthroughs in problem-solving, they come at a steep price. Gemini 4 Carbon costs 15 times more than GPT-6.1 Sol. Test details reveal that Gemini 4 Carbon ignores time limits during evaluation and rarely gives up on complex tasks, a strategy that likely drives up computational resource consumption.

Sources & editorial note

This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.

  1. the-decoder.comSource published

Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.

Related reading

All stories
88Z.ai