We study the problem of searching for a target at some unknown location in \(\mathbb {R}^d\) when additional information regarding the position of the target is available in the form of predictions. In our setting, predictions come as approximate distances to the target: for each point \(p\in \mathbb {R}^d\) that the searcher visits, we obtain a value \(\lambda (p)\) such that \(|p\boldsymbol{t}|\le \lambda (p) \le c\cdot |p\boldsymbol{t}|\) , where \(c\ge 1\) is a fixed constant, \(\boldsymbol{t}\) is the position of the target, and \(|p\boldsymbol{t}|\) is the Euclidean distance of p to \(\boldsymbol{t}\) . The cost of the search is the length of the path followed by the searcher. Our main positive result is a strategy that achieves \((12c)^{d+1}\) -competitive ratio, even when the constant c is unknown. We also give a lower bound of roughly \((c/16)^{d-1}\) on the competitive ratio of any search strategy in \(\mathbb R^d\) .