The closely contested probabilities around 50% for numerous labs underscore a wide-open race for second place in Text Arena Math by December, with no dominant frontrunner emerging from recent model releases. Google leads the pack at 34.5% implied odds thanks to its established scaling advantages and math-focused fine-tuning, while Anthropic, Mistral, and several Chinese labs like DeepSeek and StepFun cluster in the mid-20s amid credible reports of strong benchmark gains in reasoning chains and symbolic problem-solving. Traders are watching for late-year capability updates, new training datasets emphasizing multi-step mathematics, and competitive positioning against OpenAI’s lower 4.2% odds, as small shifts in demonstrated performance could quickly realign sentiment before year-end resolution.
Polymarketデータを参照したAI生成の実験的な要約。これは取引アドバイスではなく、このマーケットの解決方法には一切関係ありません。 · 更新日View resolved






















外部リンクに注意してください。
外部リンクに注意してください。
よくある質問