Property-based testing in practice often involves writing complex input generators, which can be hard to manually verify if they cover the desired set of inputs. Coverage types have emerged as a potential solution to this problem, an analog to refinement types where functions are annotated with a subset they must be able to generate. However, users may want to treat certain inputs as “equivalent,” where they assert that the testing behavior of two distinct inputs is the same. Coverage types do not easily extend to this use case since they enforce all distinct inputs must be generated. We propose an extension to the coverage type system that incorporates equivalence relations between values, where two values are equivalent if users do not care to differentiate between them. We also update the type checking algorithm to correctly and more efficiently handle equivalence queries, and we present preliminary results of the efficacy of this approach.