QOCO is a convex solver for Second-Order Cone Programs (SOCPs) which implements a primal-dual interior point method. It is written in C, has a BSD-3-Clause license, and is contained in qoco.
QOCOGEN is a code generator housed in qocogen which takes in an second-order cone program problem family and generates a customized C solver (called qoco_custom) for the specified problem family which implements the same algorithm as QOCO. This customized solver is library-free, only uses static memory allocation, and can be a few times faster than QOCO.
QOCO has a python interface, and can be used in CVXPY.
Check out the documentation to learn how to use QOCO and QOCOGEN in your projects.
QOCO's C code can be found here, and build instructions can be found here.
QOCO is available through PyPI by running
pip install qoco
QOCOGEN is only available through PyPI and can be installed by running
pip install qocogen
QOCO's matlab wrapper can be found here, and build instructions can be found here.
If you find QOCO useful please cite the QOCO paper as follows
@article{chari2026qoco,
title = {{QOCO}: a quadratic objective conic optimizer with custom solver generation},
author = {Chari, Govind M. and A\c{c}ıkmeşe, Beh\c{c}et},
journal = {Mathematical Programming Computation},
issn = {1867-2957},
url = {http://dx.doi.org/10.1007/s12532-026-00311-8},
doi = {10.1007/s12532-026-00311-8},
publisher = {Springer Science and Business Media LLC},
year = {2026},
month = mar,
}
If you find the GPU backend useful please cite the QOCO-GPU paper as follows
@article{chari2026qocogpu,
title = {{QOCO}-{GPU}: A Quadratic Objective Conic Optimizer with {GPU} Acceleration},
issn = {1867-2957},
url = {http://dx.doi.org/10.1007/s12532-026-00338-x},
doi = {10.1007/s12532-026-00338-x},
journal = {Mathematical Programming Computation},
publisher = {Springer Science and Business Media LLC},
author = {Chari, Govind M. and A\c{c}ıkmeşe, Beh\c{c}et},
year = {2026},
month = Sept
}
