For solving the large-scale nonlinear equation, a block Newton-Kaczmarz-type method based on a greedy criterion is proposed. The method is deterministic and adaptive and needs not compute the pseudoinverses of submatrices. It can be also regarded as a nonlinear verson of the fast deterministic block Kacmarz method. It is proved that the method will converge linearly to the nearest solution to the initial point under mild conditions. Numerical experiments are performed to illustrate that the method is more efficient at least for the tested problems.