[Purposes] Effectively share risks, reduce the shirking of responsibilities, and promote the smooth cooperation of the whole-process engineering consulting consortium. [Methods] This paper establishes a list of risk factors and an index system of risk-sharing influencing factors from the perspective of the whole-process engineering consulting consortium. The subjective and objective weights of each index are calculated by using the improved analytic hierarchy process based on the cloud model and the entropy weight method, and the combination weights are obtained by using the linear weighting method. Subsequently, an internal risk-sharing model (GRA-TOPSIS) based on grey relational analysis (GRA) coupled technique for order preference by similarity to an ideal solution (TOPSIS) is constructed to determine the risk-sharing results by ranking the comprehensive relative proximity. [Findings] The model constructed in this paper is used for a construction project in Changsha, and the most suitable risk-sharing party is selected by taking the engineering change risk as an example. The result is consistent with the actual project. Then, the solution determines a set of reasonable sharing or exclusive risk schemes, which achieves the rational sharing of internal risks within the consortium. [Conclusions] The GRA-TOPSIS model is suitable for the internal risk-sharing of the whole process engineering consulting consortium and has universality, which can provide a certain reference for the promotion of the internal risk-sharing of the consortium and the whole process engineering consulting service model.