HiGHS is a C++11-based solver for LP (linear programming), MIP (mixed-integer programming), and convex QP (quadratic programming). Its origins date to late 2016, when presolve work by Ivet Galabova was combined with dual/simplex work by Qi Huangfu and Julian Hall; the project and name matured through 2018 onward.

Primal & dual revised-simplex engines were originally implemented by Qi Huangfu and later developed further by Julian Hall. The interior-point (IPM) solver for large LPs was developed by Lukas Schork. The active-set QP solver for convex quadratic programs was developed by Michael Feldmeier. MIP (branch-and-cut) capabilities include more recent contributions from Leona Gottwald. Presolve, crash procedures, and general engineering work were led by Ivet Galabova (presolve) and Julian Hall (engineering & integration).

HiGHS is written with minimal external dependencies (CMake build, no heavy third-party stack), ships as a library and an executable (bin/highs), and exposes language bindings for Python (highspy), C, C#, Fortran, Julia, and others. That makes it easy to call from modeling stacks and numerical ecosystems.

HiGHS is published under the MIT license and provides prebuilt binaries and language packages for common platforms.