cma
CMA-ES, Covariance Matrix Adaptation Evolution Strategy for non-linear numerical optimization in Python
Decision gist · record as of 2026-08-14
Yes. CMA-ES is a well-maintained, production-stable algorithm with no known vulnerabilities, low install friction, and permissive licensing. Install it if you need derivative-free optimization for difficult non-convex problems in moderate dimensions; skip it if you have gradients available or are optimizing in very small dimensions where faster methods exist.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Low friction install with only numpy as a runtime dependency.
- Active maintenance with recent releases and a stable codebase.
License · maintenance · safety
BSD-3-Clause (permissive) — BSD-3-Clause is a permissive license allowing commercial and private use with minimal restrictions.
last release 2026-02-25 (170 days) · last repo commit 2026-08-05 · 1,344 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 1,813,205 downloads/mo, #3,527 on PyPI
Alternatives
Verify before relying
pip install cma
import cma
es = cma.CMAEvolutionStrategy(8 * [0], 0.5)
xopt, es = cma.fmin2(cma.ff.rosen, 8 * [0], 0.5)- Whether the package supports Python 3.10+ or has specific version constraints beyond 'Python 3'
What it is and what it does
CMA-ES (Covariance Matrix Adaptation Evolution Strategy) is a randomized, derivative-free optimization algorithm implemented in Python for solving difficult continuous and mixed-integer optimization problems. It is designed for objective functions where gradients are unavailable or unhelpful, the search space has dependent variables, and the landscape may be non-convex, ill-conditioned, multi-modal, or noisy. The package provides two independent implementations via the CMAEvolutionStrategy class and a basic purecma.CMAES variant.
The algorithm is most effective for problems with search space dimensions between five and a few hundred, requiring at least around 100 times the dimension in function evaluations. CMA-ES handles bound constraints, linear and nonlinear constraints, noise, and integer variables for mixed-integer problems. It offers both a direct optimization interface (fmin2) and an ask-and-tell interface for fine-grained control over the iteration loop. The package depends only on numpy for array operations.
Use it for
- Optimize expensive black-box functions where gradients are unavailable or unreliable
- Solve non-convex or multi-modal optimization problems in 5-100+ dimensions
- Handle constrained optimization with bounds or nonlinear constraints
- Tune hyperparameters or design parameters in machine learning or engineering workflows
- Optimize noisy objective functions where derivative-based methods fail
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes.
CMA-ES is a well-maintained, production-stable algorithm with no known vulnerabilities, low install friction, and permissive licensing. Install it if you need derivative-free optimization for difficult non-convex problems in moderate dimensions; skip it if you have gradients available or are optimizing in very small dimensions where faster methods exist.
Install
cma on PyPI
Before you install
Low friction install with only numpy as a runtime dependency. Active maintenance with recent releases and a stable codebase.
License in practice
BSD-3-Clause is a permissive license allowing commercial and private use with minimal restrictions.
Quickstart
pip install cma
import cma
es = cma.CMAEvolutionStrategy(8 * [0], 0.5)
xopt, es = cma.fmin2(cma.ff.rosen, 8 * [0], 0.5)Verify before relying
- Whether the package supports Python 3.10+ or has specific version constraints beyond 'Python 3'
Package facts
| License | BSD-3-Clause permissive |
| Python support | Not specified |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 1 packagenumpy |
| Maintenance | Actively maintained 170 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 1,813,205 / month, #3,527 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 5 - Production/StableEnvironment :: ConsoleFramework :: IPythonFramework :: JupyterIntended Audience :: EducationIntended Audience :: Other AudienceIntended Audience :: Science/ResearchOperating System :: OS IndependentProgramming Language :: Python :: 2.7Programming Language :: Python :: 3Topic :: Scientific/EngineeringTopic :: Scientific/Engineering :: Artificial IntelligenceTopic :: Scientific/Engineering :: Mathematics |
Evidence: cma-4.4.4-py3-none-any.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 › “CMA-ES algorithm”
- cmaCMA-ES is a derivative-free numerical optimization algorithm for…
- deapDEAP is an evolutionary computation framework for implementing…
- cmaesProvides CMA-ES (Covariance Matrix Adaptation Evolution Strategy)…
Give your agent the search over MCP, or paste the wish link into any chat.
More Scientific/Engineering packages
NumPy provides an N-dimensional array object and a comprehensive suite of mathematical, linear algebra, Fourier transform, and random number functions for scientific computing in Python.
pandas provides fast, flexible data structures (Series and DataFrame) for loading, cleaning, transforming, and analyzing labeled or relational data in Python.
scipy provides numerical algorithms for mathematics, science, and engineering—including optimization, integration, linear algebra, Fourier transforms, signal and image processing, and ODE solvers—built on numpy arrays.
scikit-learn provides a comprehensive Python library for supervised and unsupervised machine learning, including classification, regression, clustering, dimensionality reduction, and model evaluation tools built on NumPy and SciPy.
Install it if you need to train, evaluate, or deploy supervised or unsupervised learning models.
dill extends Python's pickle module to serialize and deserialize a much wider range of Python objects, including functions, lambdas, classes, and interpreter sessions, to byte streams for storage or network transmission.
Multiprocess is an enhanced fork of Python's standard multiprocessing library that uses dill for better serialization, allowing you to spawn processes with a threading-like API and share complex objects between them.
Install it if you use multiprocessing and encounter pickle serialization limits with lambdas or complex objects.
See also cmaes · directsearch · deap · ropwr · bayesian-optimization · scikit-optimize · Mosek · nevergrad · nvidia-modelopt · nvidia-cusolver-cu11