Traders see a tight race for third place on the style-controlled Chatbot Arena leaderboard by month-end, with Meta and Alibaba holding the clearest edges at roughly 32% and 31% implied odds due to their strong recent model releases and consistent mid-tier positioning once length and formatting biases are removed. Alibaba’s Qwen series and Meta’s ongoing iterations have shown competitive substance scores that hold up under the adjusted metrics, while Google’s Gemini variants trail at 13% amid slower visible progress on style-normalized benchmarks. Lower odds for OpenAI and others reflect persistent gaps in controlled evaluations, underscoring how developer focus on concise, high-signal outputs and targeted fine-tuning can shift standings quickly in the final weeks.