The current state-of-the-art in iterative optimization algorithms for differentiable cost functions is scattered throughout the literature, which hinders their comparison for the specific program at hand. Depending on the research area, theoretical optimal parameters and convergence rates are available in different formulations. Consequently, this toolbox aims at providing a benchmarking software for the various gradient-descent-based algorithms and implements functions to return the optimal parameters whenever possible. Researchers can focus on the development of new algorithms and test them against the ones present in the literature and made them available under a common framework. The structure of the software is tailored such that novel contributions can be easily added through the design of a function implementing a single step of the algorithm. The ease of usage is illustrated by examples in the literature.