The article discusses the CUTE environment, a versatile testing framework for nonlinear optimization algorithms. It describes a collection of test problems in the standard input format (SIF) used by the LANCELOT software package. These problems are classified to help identify specific problem types, and tools are provided to manage this database. The environment includes tools for interfacing with existing optimization packages like MINOS and OSL, as well as for building interfaces with new algorithms. The test problems cover a wide range of applications, including academic and real-world scenarios, and include both constrained and unconstrained examples. The classification system allows for efficient selection and management of problems based on their characteristics. The article also describes the tools for adding new problems, managing classification databases, and interfacing with existing optimization packages. The CUTE environment is designed to be useful for researchers in developing optimization software, and the tools are available for download via anonymous ftp. The article provides detailed information on how to obtain and install the CUTE environment, as well as how to use the provided tools for testing and evaluating optimization algorithms.The article discusses the CUTE environment, a versatile testing framework for nonlinear optimization algorithms. It describes a collection of test problems in the standard input format (SIF) used by the LANCELOT software package. These problems are classified to help identify specific problem types, and tools are provided to manage this database. The environment includes tools for interfacing with existing optimization packages like MINOS and OSL, as well as for building interfaces with new algorithms. The test problems cover a wide range of applications, including academic and real-world scenarios, and include both constrained and unconstrained examples. The classification system allows for efficient selection and management of problems based on their characteristics. The article also describes the tools for adding new problems, managing classification databases, and interfacing with existing optimization packages. The CUTE environment is designed to be useful for researchers in developing optimization software, and the tools are available for download via anonymous ftp. The article provides detailed information on how to obtain and install the CUTE environment, as well as how to use the provided tools for testing and evaluating optimization algorithms.