OpenAI holds the strongest position for second place on the Code Arena WebDev leaderboard through mid-October 2026, with its GPT-6 Astra-max and GPT-6.1 Sol models scoring in the 1760–1780 range, just behind Anthropic’s leading Claude Opus 5.5 and Sonnet 5.5 variants. Consistent performance across front-end tasks, agentic workflows, and multi-step reasoning has kept OpenAI ahead of Google’s Gemini 4 series and other labs. Recent model updates from OpenAI have narrowed gaps without major slips, while competitors like Moonshot, Meta, and Alibaba trail by larger margins with fewer recent gains. Traders assign lower odds to rapid shifts before November’s end, though new releases or benchmark volatility could still alter the order.