Gemma 4 reasoning flag lifts accuracy 48 points, cuts cost 6-8x
California – October 31, 2026 -- Switching off Gemma 4's default reasoning with a single flag raised accuracy by up to 48 points, cut cost 6x to 8x and made a correct answer up to 19x faster on short verifiable tasks, according to a 27-model local benchmark run on a Strix Halo machine with a power meter attached. The 12B and 26B variants rose from about 52% to 100% on the easy suite.
Each model answered 190 checkable tasks twice, under two different prompts. Electricity was priced at California summer rates of 34 cents per kilowatt-hour off-peak and 58 cents on-peak.
Gemma 4 reasons on 95 of 95 tasks even when told not to
Asked for a one-line answer with no explanation, Gemma 4 emitted reasoning on every task in both prompt conditions. On a 4,096-token budget, the 12B and 26B spent the entire allowance thinking and returned nothing on roughly half the questions.