proxsuite
Quadratic Programming Solver for Robotics and beyond.
Decision gist · record as of 2026-08-14
Yes, if you need a robust, actively maintained QP solver with good performance on hard problems and integration into major optimization ecosystems. The permissive BSD-2-Clause license, medium install friction, and zero known vulnerabilities make it a safe choice. Install with caution only if your platform is not among the prebuilt wheel targets (Linux x86_64/aarch64, macOS Intel/ARM, Windows).AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python 3.9 or later; prebuilt wheels available for common platforms, but compilation from source may be needed on unsupported architectures.
- Medium install friction due to compiled C++ backend with prebuilt wheels for Python 3.9–3.14 on Linux, macOS (Intel and ARM), and Windows.
- Active maintenance with last release 95 days ago; no known vulnerabilities.
License · maintenance · safety
BSD-2-Clause (permissive) — BSD-2-Clause permissive license allows commercial and private use with minimal restrictions; attribution required.
last release 2026-05-11 (95 days) · last repo commit 2026-05-12
0 known vulnerabilities (OSV.dev, 2026-08-14) · 102,765 downloads/mo, #12,850 on PyPI
Alternatives
Verify before relying
pip install proxsuite
import proxsuite
import numpy as np
H = np.eye(2)
g = np.array([1., 2.])
qp = proxsuite.proxqp.dense.QP(H, g)
qp.solve()- Whether the package supports Julia bindings as described in the excerpt, or if those are separate.
- Performance characteristics on specific problem sizes or sparsity patterns beyond the benchmarks mentioned.
- Availability and maturity of the QPLayer differentiable QP layer feature.
What it is and what it does
ProxSuite is a collection of numerical solvers for quadratic programs (QPs), linear programs (LPs), and related optimization problems, built on primal-dual proximal algorithms. It targets robotics applications but is domain-agnostic and can solve dense, sparse, or matrix-free problems. The package is a C++ template library with Python bindings, offering warm-starting, batch solving, handling of nonconvex QPs, and infeasibility detection.
The core solver, ProxQP, handles standard QP form with equality and inequality constraints. It is integrated into CVXPY, CasADi, and TSID, and is actively maintained by research groups at Inria and École Normale Supérieure de Paris. Installation is straightforward via pip, conda, or brew on Linux, macOS, and Windows; runtime dependencies are numpy and scipy.
Use it for
- Solve quadratic programs in robotics applications, such as inverse dynamics with contacts via TSID.
- Use as a backend solver within CVXPY or CasADi for convex optimization modeling.
- Handle large-scale or sparse QPs where specialized matrix factorization backends improve performance.
- Solve batches of QPs in parallel for parameter-sweep or sensitivity analysis workflows.
- Detect and handle infeasible QPs or find the closest feasible solution when problems are ill-posed.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you need a robust, actively maintained QP solver with good performance on hard problems and integration into major optimization ecosystems.
The permissive BSD-2-Clause license, medium install friction, and zero known vulnerabilities make it a safe choice. Install with caution only if your platform is not among the prebuilt wheel targets (Linux x86_64/aarch64, macOS Intel/ARM, Windows).
Install
proxsuite on PyPI
Before you install
Medium install friction due to compiled C++ backend with prebuilt wheels for Python 3.9–3.14 on Linux, macOS (Intel and ARM), and Windows. Active maintenance with last release 95 days ago; no known vulnerabilities.
Requires Python 3.9 or later; prebuilt wheels available for common platforms, but compilation from source may be needed on unsupported architectures.
License in practice
BSD-2-Clause permissive license allows commercial and private use with minimal restrictions; attribution required.
Quickstart
pip install proxsuite
import proxsuite
import numpy as np
H = np.eye(2)
g = np.array([1., 2.])
qp = proxsuite.proxqp.dense.QP(H, g)
qp.solve()Verify before relying
- Whether the package supports Julia bindings as described in the excerpt, or if those are separate.
- Performance characteristics on specific problem sizes or sparsity patterns beyond the benchmarks mentioned.
- Availability and maturity of the QPLayer differentiable QP layer feature.
Package facts
| License | BSD-2-Clause permissive |
| Python support | Supports the current Python release >=3.9 |
| Install friction | Medium. Platform-specific wheel |
| Runtime dependencies | 3 packagescmeelnumpyscipy |
| Maintenance | Actively maintained 95 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 102,765 / month, #12,850 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
Evidence: proxsuite-0.7.3-2-cp310-cp310-macosx_10_9_x86_64.whl; proxsuite-0.7.3-2-cp310-cp310-macosx_11_0_arm64.whl; proxsuite-0.7.3-2-cp310-cp310-manylinux_2_28_aarch64.whl; proxsuite-0.7.3-2-cp310-cp310-manylinux_2_28_x86_64.whl; proxsuite-0.7.3-2-cp311-cp311-macosx_10_9_x86_64.whl; proxsuite-0.7.3-2-cp311-cp311-macosx_11_0_arm64.whl; proxsuite-0.7.3-2-cp311-cp311-manylinux_2_28_aarch64.whl; proxsuite-0.7.3-2-cp311-cp311-manylinux_2_28_x86_64.whl; proxsuite-0.7.3-2-cp312-cp312-macosx_10_9_x86_64.whl; proxsuite-0.7.3-2-cp312-cp312-macosx_11_0_arm64.whl; proxsuite-0.7.3-2-cp312-cp312-manylinux_2_28_aarch64.whl; proxsuite-0.7.3-2-cp312-cp312-manylinux_2_28_x86_64.whl; proxsuite-0.7.3-2-cp313-cp313-macosx_10_9_x86_64.whl; proxsuite-0.7.3-2-cp313-cp313-macosx_11_0_arm64.whl; proxsuite-0.7.3-2-cp313-cp313-manylinux_2_28_aarch64.whl; proxsuite-0.7.3-2-cp313-cp313-manylinux_2_28_x86_64.whl; proxsuite-0.7.3-2-cp314-cp314-macosx_10_9_x86_64.whl; proxsuite-0.7.3-2-cp314-cp314-macosx_11_0_arm64.whl; proxsuite-0.7.3-2-cp314-cp314-manylinux_2_28_aarch64.whl; proxsuite-0.7.3-2-cp314-cp314-manylinux_2_28_x86_64.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “QP solver for robotics”
- proxsuiteProxSuite is a numerically robust quadratic programming solver that…
- daqpDAQP is a dual active-set solver for convex quadratic programs,…
- osqpOSQP is a Python wrapper for the Operator Splitting Quadratic Program…
Give your agent the search over MCP, or paste the wish link into any chat.
More Mathematics packages
NetworkX provides data structures and algorithms for creating, analyzing, and manipulating graphs and networks, supporting everything from simple undirected graphs to complex directed and weighted networks.
kiwisolver is a Python binding to a fast C++ implementation of the Cassowary constraint solver, enabling you to solve systems of linear constraints and inequalities.
Install it if you need to solve constraint systems; skip it if you only need simple linear algebra.
SymPy is a Python library for symbolic mathematics, performing algebraic manipulation, calculus, equation solving, and mathematical expression simplification without numerical approximation.
ContourPy calculates contours of 2D quadrilateral grids using C++11 algorithms wrapped in Python, offering serial and multithreaded implementations without requiring Matplotlib as a dependency.
PyTorch provides GPU-accelerated tensor computation and automatic differentiation for building and training deep neural networks in Python.
onnxruntime loads and executes Open Neural Network Exchange (ONNX) models with a focus on inference performance across CPUs and accelerators.
Install it if you have ONNX models to run in production or development.
See also qpsolvers · osqp · quadprog · daqp · cvxopt · cvxpy · directsearch · Mosek · optlang