**Tight competition among frontier labs defines trader sentiment for the second-best Text Arena Math performer by end of December, with market-implied odds clustered near 50% across Google, Anthropic, Alibaba, DeepSeek, OpenAI, and others.** Recent Gemini 4 Argon releases have propelled Google to the top of the lmarena.ai Math leaderboard with Elo scores exceeding 1525, while Anthropic’s Claude Opus and Fable variants remain within striking distance at 1522–1532. Frequent model drops, including Chinese labs’ Qwen3.8-max and GLM-5.3 iterations plus OpenAI’s internal math-focused advances, have driven 17 leaderboard shifts in recent months. Traders price in continued volatility from upcoming releases, benchmark updates, and capability demonstrations over the next quarter, with no single lab holding a durable edge in this dynamic category.