COBYQA (Constrained Optimization BY Quadratic Approximations) is a derivative-free trust-region SQP optimization solver for constrained optimization using quadratic approximations.

It is currently implemented in Python.

  • A complete documentation is available.
  • The source files can be downloaded from GitHub.
  • Compiled versions can be installed from PyPI.
  • A detailed description of the algorithm and its implementation is given in my PhD thesis.