Anthropic's Claude Sonnet 5.5 model has secured third place on the Arena.ai Code Arena WebDev leaderboard. The model achieved a score of 1786 points using xHigh reasoning. It currently trails the second-place GPT-6 Astra Max from OpenAI by approximately six points.
The top position on the leaderboard is held by Claude Opus 5.5 Max, which scored between 1818 and 1820 points. The ranking is based on community votes where users compare model outputs for front-end web development tasks involving HTML and React.
Launched around September 28, 2026, Claude Sonnet 5.5 is positioned as a mid-tier offering. Its pricing is set at $2 per million input tokens and $10 per million output tokens, which is half the cost of the flagship Opus 5.5 line.
Beyond the leaderboard, Sonnet 5.5 outperformed Opus 5.5 on the Terminal-Bench 4.0 coding benchmark, scoring 70.6% compared to 66.4%. As of late September 2026, the Arena.ai leaderboard had aggregated over 800,000 votes across more than 100 models.
