Google reports 65.2% for Gemini 4 Carbon on Terminal-Bench 4.0, citing an Artificial Analysis run.
Tasks: Artificial Analysis run, labeled Terminal-Bench 4.0 (AA) by Google.
Setting: Google says its results use the Gemini API at the highest thinking setting unless noted; no numeric budget is stated.
Limits: Provider-reported: Google publishes the result for its own model; it has not been independently reproduced here. Google says this score comes from a third-party leaderboard; that leaderboard was not opened here, so its run date, harness and settings are unverified. Google says the scores are as of October 2026 for the model it calls Gemini 4 Carbon; equivalence to the build now served on every surface is not established. Results under the same benchmark name can be shown together as reported scores, but unmatched settings and missing methods rule out a controlled ranking. A third-party run of Terminal-Bench 4.0 can use a different harness from a model maker's own run.
Google DeepMind · Google DeepMind · Google DeepMind
