PIQP is an Proximal Interior Point Quadratic Programming solver, which can solve dense and sparse quadratic programs described in described in Schwan, Jiang, Kuhn, and Jones (2023) (https://arxiv.org/abs/2304.00290). Combining an infeasible interior point method with the proximal method of multipliers, the algorithm can handle ill-conditioned convex QP problems without the need for linear independence of the constraints. The solver is written in header only 'C++ 14' leveraging the Eigen library for vectorized linear algebra. For small dense problems, vectorized instructions and cache locality can be exploited more efficiently. Allocation free problem updates and re-solves are also provided.