I built a Python package called gridoptim that performs fast parallel grid search optimisation.

SciPy provides scipy.optimize.brute, but it becomes extremely slow for large grids because it evaluates every point sequentially in Python.

gridoptim parallelizes the search and evaluates the objective efficiently.

In a benchmark on a 4D grid (64^4 = 16,777,216 evaluations):

gridoptim: ~1.2 seconds scipy.brute: ~247 seconds

≈ 205× speedup on a 12-core machine.

The project is open source:

GitHub: https://github.com/Halfblood-Prince/gridoptim PyPI: https://pypi.org/project/gridoptim/