In mice that exercise, cutting carbs down to just 1% of their diet — but not 10% — turns on ketosis, a state where the body burns fat for fuel, and this only happens when protein levels are kept the same.

What we've found so far is that in exercise-trained mice, a 1% carbohydrate diet appears to induce ketosis, while a 10% carbohydrate diet does not, when protein intake is matched [1]. The evidence we've reviewed supports this pattern, with 14.0 assessments backing it and none contradicting it [1]. Our analysis of the available research suggests that carbohydrate levels play a key role in triggering ketosis in mice that are physically active. Ketosis is a metabolic state where the body shifts from using glucose (from carbs) to using fat as its main fuel source. In the studies we examined, this shift only occurred when carbohydrates were reduced to just 1% of the diet [1]. When carbs were increased to 10%, even with the same protein intake, ketosis did not occur [1]. This suggests that the threshold for inducing ketosis in exercised mice may be quite low in terms of carbohydrate intake. We only looked at one assertion so far, but it has been supported consistently across 14.0 evaluations [1]. Still, our current analysis is limited to this single claim and applies specifically to mice that exercise under protein-matched conditions. We cannot say whether the same threshold would apply in other animals or in humans, or under different dietary or activity conditions. What this might mean for future research is that small changes in carbohydrate intake—especially at very low levels—could have meaningful effects on metabolic states like ketosis, at least in active mice. But we need more data to understand how these findings might extend beyond this specific context. For now, what we can say is that the evidence we've reviewed leans toward a clear difference between 1% and 10% carbohydrate diets in triggering ketosis in exercised mice when protein is held constant [1].