Traders view the race for second-best in text-based math arenas by October's end as tightly contested, with several labs clustered near 50% implied probability amid rapid advances in large language model reasoning. Strong showings on math benchmarks from Chinese developers like DeepSeek and Alibaba, alongside steady progress at Google and OpenAI, have kept the field open without a dominant contender. Key differentiators include specialized fine-tuning on complex problem sets, chain-of-thought techniques, and efficient scaling methods that boost accuracy on competition-level tasks. Upcoming model releases and benchmark updates could quickly reshape the ordering as labs compete in this high-stakes capability area.