Recent releases from OpenAI and Anthropic have driven sharp gains on arena.ai's Code Arena WebDev leaderboard, where human-voted Elo ratings evaluate agentic coding and web development tasks. As of late September 2026, OpenAI's GPT-6 Astra Max leads at approximately 1800 Elo, ahead of Anthropic's Claude Fable 5.1 Max near 1758 and other Claude Opus variants in the mid-1600s. These scores reflect rapid progress in multi-step reasoning and tool use, outpacing earlier 2026 peaks around 1570. Competitive pressure from Alibaba's Qwen and Meta models adds volatility, while Q4 model updates or fine-tunes from frontier labs remain the primary catalysts that could push additional entries past higher thresholds before year-end resolution. Trader sentiment tracks these leaderboard shifts closely given the direct resolution link.