Cognition scores Gemini 4 Carbon 40% and Argon 26.3% on FrontierCode 2.0

Cognition says Gemini 4 Carbon scores 40% on its FrontierCode 2.0 coding benchmark, at $4.04 per task.

It places behind Claude Opus 5.5, GPT-6 Astra and Claude Sonnet 5.5, and Cognition describes it as near state of the art.

Cognition says Gemini 4 Argon scores 26.3% at $6.41 per task. That puts it ahead of Muse Wonder 1.0 but behind Claude Haiku 5.5 and Grok 4.7.

Cognition describes FrontierCode 2.0 as a benchmark of a model's ability to go beyond the spec. Full scores are posted on the FrontierCode leaderboard at cognition.com/frontiercode.