Google reports 53.4% for Gemini 4 Hydrogen on the Multimodal Knowledge Agent (MKA) benchmark, which Google developed with Artificial Analysis, from its own runs.
Tasks: Climb WSR accuracy, as Google labels the metric.
Setting: Google says its results use the Gemini API at the highest thinking setting (xhigh for Hydrogen) 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 co-developed MKA with Artificial Analysis and introduced it on November 10, 2026, two days before reporting Hydrogen's result on it. Google says the scores are as of November 2026 for the model it calls Gemini 4 Hydrogen; 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.
Google DeepMind · Google DeepMind · Google DeepMind · Google DeepMind · Google · Google
