Google's Gemini 4 Argon currently leads Text Arena math leaderboards with the highest Elo ratings from blind votes on math prompts, giving traders a slight edge in implied probability at 49.5% versus Anthropic's 36.5%. Recent releases and optimizations in Google's Gemini 4 series have strengthened its position in multi-step reasoning and competition-style problems, outpacing Anthropic's Claude Opus 5 variants that remain statistically close. Other labs like Z.ai, Alibaba's Qwen, and Meta trail further behind on arena metrics, while OpenAI models sit lower despite strong benchmark scores elsewhere. Rapid iteration in large language models means new releases or fine-tunes before end of January could shift the outcome, with traders watching for updates that improve math-specific capabilities like chain-of-thought accuracy and consistency.
Experimental AI-generated summary referencing Polymarket data. This is not trading advice and plays no role in how this market resolves. · Updated1H6H1D7D30DALL
View resolved






















Beware of external links.
Beware of external links.
Frequently Asked Questions