systems, schema evolution has been explored extensively: [16] surveys practical challenges of schema evolution and migration, MigCast [5] estimates the cost and performance implications of data-model changes, [3] introduces a formal taxonomy of schema modifications, [2] proposes generic evolution techniques applicable across both NoSQL and relational models, [4] presents a categorical framework for migrations between NoSQL models, Darwin [15] offers a platform for multi-model data evolution, and [7] defines basic operations for both evolution and migration of multi-model data. Although these works highlight the importance of structural adaptation, none address query-driven schema changes in multi-model settings. The natural solution – automatically suggesting or adapting data representations in multi-model settings – introduces several challenges. (1) The search space of possible mappings grows exponentially with the number of entities and their possible embeddings, inlinings, or placements across models. (2) Multimodel query workloads may combine different operator families (relational, document, graph), and the impact of a representation change often propagates across model boundaries. (3) Evaluating each candidate mapping requires estimating performance for heterogeneous operators, which is difficult without a unified model of query-plan behavior. To address these challenges, we introduce MM-mapsearch, a tool that analyses multi-model workloads and recommends performance-oriented schema or mapping changes. Its main contributions are: Modern database systems offer several alternative representations of the same data, both within one model and across multiple models. These choices have a strong influence on performance. But manually identifying an efficient mapping is challenging, especially in multi-model environments that combine heterogeneous operators. In this paper, we present MM-mapsearch, an automatic advisor for selecting workload-aware multi-model mappings. A simplified categorical schema defines the search space and enables uniform reasoning about structural alternatives. MM-mapsearch analyses a weighted workload, explores feasible schema variants using Monte Carlo Tree Search, and estimates their performance with a plan-structured neural predictor that generalizes across relational, document, and graph systems.