**OpenAI holds a commanding 90% implied probability as the second-ranked lab on the Code Arena WebDev leaderboard because Anthropic’s Claude Opus 5.5 Max and related models currently occupy the top lab position with scores around 1815, while OpenAI’s GPT-6 Astra Max and GPT-6.1 Sol keep the company firmly in second at approximately 1788.** Recent releases, including OpenAI’s GPT-6.1 Terra on October 26–27, which reached #14 overall with strong cost efficiency, have reinforced this positioning without closing the gap to Anthropic. Other labs trail further behind—Google, Alibaba, Moonshot, and Meta sit in the 1650–1680 range—creating a wide enough margin that no meaningful shift is expected before the October 31 resolution. Trader consensus, backed by real capital on Polymarket, treats the current lab ordering as stable through month-end absent any surprise model drops.