We linked seven-year-old children's personality (n = 406), as rated by both teachers and parents, to a wide array of cognitive ability measures. Besides (i) providing descriptive data on the associations between childhood personality and cognitive ability, we (ii) investigated the benefits of having multiple informants provide personality ratings and (iii) examined the recent proposition that the general factor of intelligence be partialled out of associations between personality and narrower domains of intelligence. In a regression model, the shared variance of teacher and parent ratings of personality explained 14% of the variance of cognitive ability. Both teacher and parent ratings of Openness to Experience (O) were positively associated with cognitive ability, and both explained unique variance in cognitive ability. Moreover, the associations were stronger the stronger the inter-rater agreement on O. When the unique variances of each perspective and the moderating effect of inter-rater agreement were added to the aforementioned regression model, personality explained 18% of the variance of cognitive ability. Controlling for the general factor of intelligence caused the correlations between personality and performance on the specific cognitive ability measures to diminish to near zero. Copyright © 2011 John Wiley & Sons, Ltd.