The six-month gap between leaderboard rankings and spending suggests that human preference is a leading indicator for adoption. The current market lead follows Claude Opus 4.7 Thinking rankings that currently dominate the field, even as OpenAI's Codex usage surged 5x since January to 3 million weekly developers.
This shift signals a move toward model specialization. Anthropic has seen its performance lead translate into enterprise growth, while OpenAI extends GPT-5.5's agentic coding performance (AI that autonomously writes and iterates on code). Arena.ai indicates that the next competitive frontier will be autonomous coding agents.
According to @tryramp, @AnthropicAI just overtook OpenAI in business customers (34.4% vs 32.3% this week). In the Text Arena, that flip happened in Q4 2025. Real-world signal led enterprise adoption by ~6 months. But the picture shifts fast: Codex crossed 3M+ weekly developers in April, up 5x since January (per @OpenAI). Coding agents are their own contest. More from Arena shortly.
More like this
