The field of database technology has expanded to include diverse data models, leading to the exploration of the polystore concept, which aims to provide unified management for heterogeneous data. However, the development of multi-model query languages, which serve as the primary interface for polystores, remains a significant challenge. Existing multi-model query languages face limitations, such as restricted expressiveness in query representation and inflexibility imposed by predefined data models. To address these problems, this paper introduces the Equivalent Combinations of Query Languages (ECQL) as a holistic representation of multi-model queries. We first establish foundational abstractions for the query results across various data models. Subsequently, we construct a novel multi-model query language called ECQL that naturally combines sub-queries of different data models through expression interfaces based on these abstractions. The expressiveness of ECQL lies in its capability to support both single-model and cross-model query operations. Comparative evaluations against existing multi-model query languages on polystore benchmarks underscore the succinctness of ECQL. Finally, we illustrate the extensibility of ECQL by demonstrating its compatibility with the incorporation of query languages that are not initially included in ECQL’s grammar.