The paper introduces MiniZinc, a new CP (Constraint Programming) modelling language designed to be simple yet expressive, suitable for a wide range of solvers. It also proposes FlatZinc, a low-level solver-input language, and a straightforward translation from MiniZinc to FlatZinc that preserves all solver-supported global constraints. This translation allows solver writers to support MiniZinc with minimal effort by providing a simple FlatZinc front-end and combining it with an existing translator. The authors argue that a standard CP modelling language will encourage experimentation and comparisons between different solvers, making it easier for modellers to switch between solvers and for solver writers to benchmark their solvers. MiniZinc is described as a medium-level declarative language that supports sets, arrays, user-defined predicates, and automatic coercions, while being low-level enough to be easily mapped onto most solvers. The paper concludes with a discussion of supporting tools, experimental results, related work, and future directions.The paper introduces MiniZinc, a new CP (Constraint Programming) modelling language designed to be simple yet expressive, suitable for a wide range of solvers. It also proposes FlatZinc, a low-level solver-input language, and a straightforward translation from MiniZinc to FlatZinc that preserves all solver-supported global constraints. This translation allows solver writers to support MiniZinc with minimal effort by providing a simple FlatZinc front-end and combining it with an existing translator. The authors argue that a standard CP modelling language will encourage experimentation and comparisons between different solvers, making it easier for modellers to switch between solvers and for solver writers to benchmark their solvers. MiniZinc is described as a medium-level declarative language that supports sets, arrays, user-defined predicates, and automatic coercions, while being low-level enough to be easily mapped onto most solvers. The paper concludes with a discussion of supporting tools, experimental results, related work, and future directions.