Pre-release

This release is a pre-release and may not be stable for production use.

Rapid Optimization Library (ROL)

Rapid Optimization Library

ROL (as in rock and roll) is a high-performance C++ library for numerical optimization. ROL brings an extensive collection of state-of-the-art optimization algorithms to virtually any application. Its programming interface supports any computational hardware, including heterogeneous many-core systems with digital and analog accelerators. ROL has been used with great success for optimal control, optimal design, inverse problems, image processing and mesh optimization, in application areas including geophysics, structural dynamics, fluid dynamics, electromagnetics, quantum computing, hypersonics and geospatial imaging.

For additional details, see https://rol.sandia.gov.

Feature highlights:

Vector abstractions and matrix-free interface for universal applicability
Modern algorithms for unconstrained and constrained optimization
Easy-to-use methods for stochastic and risk-aware optimization
Fast and robust algorithms for nonsmooth optimization
Trust-region methods for inexact and adaptive computations
PDE-OPT application development kit for PDE-constrained optimization
Interfaces and algorithms for optimal experimental design

Getting Started

ROL is a C++ library with a cmake build system. There are minimal third-party library requirements consisting of BLAS and LAPACK which also require a suitable Fortran compiler on the system. On Linux-based systems with an apt package manager, the dependencies may be installed with

apt install cmake gfortran libopenblas-dev liblapack3

Some typical configure scripts may be found in the .github/workflows/ directory. An in-source release build including ROL's examples and tests may be configured from within the source directory with

cmake -D CMAKE_BUILD_TYPE:STRING=RELEASE \
      -D ENABLE_EXAMPLES:BOOL=ON \
      -D ENABLE_TESTS:BOOL=ON \
      -B build \
      -D CMAKE_INSTALL_PREFIX=build/install \
      .

After a successful configure, ROL is built by then changing to the build directory and running make

cd build
make

One can then install it

make install

Building against ROL

Assuming we have a directory called 'roltest' which contains files we would like to run using ROL. In /dir/to/roltest, create a CMakeLists.txt with the following:

cmake_minimum_required(VERSION 3.10)  
project(test)
set(CMAKE_CXX_STANDARD 17) 

enable_language(C)
enable_language(CXX)

message(STATUS "ROL directory is: ${ROL_DIR}/include")
include_directories("${ROL_DIR}/include") #one down for all files
find_library(ROL_LIB NAMES rol librol PATHS ${ROL_DIR}/lib NO_DEFAULT_PATH)
if(ROL_LIB)
  message(STATUS "We found ROL!")
else()
  message(FATAL_ERROR "ROL library not found!")
endif()

add_library(rol::rol UNKNOWN IMPORTED)
set_target_properties(rol::rol PROPERTIES IMPORTED_LOCATION "${ROL_LIB}")
add_executable(ROLtest mytest.cpp)
target_link_libraries(ROLtest PRIVATE rol::rol) 

add_executable(ROLtest mytest.cpp)

The file in mytest then simply tests that it can find a basic ROL type:

#include "ROL_StdVector.hpp"
typedef double RealT;

int main() {
  int dim = 100; 

  ROL::StdVector<RealT> x(dim);
  x.randomize(-1.0, 1.0);
  std::cout << "What's the norm of x? " << x.norm() << std::endl;
}

Compiling and executing should then be a matter of:

$ cmake . -DROL_DIR=/Users/../path/to/rol/build/install
$ make
$ ./ROLtest
What's the norm of x? "some random number"

Python Interface

The python pip package rol-python is the official python interface to ROL. It is available at https://pypi.org/project/rol-python/ and can be installed via

pip install rol-python

Copyright and License

See COPYRIGHT and LICENSE.

Questions?

Contact team or developers:

ROL Team (GitHub handle: @sandialabs/rol)
Drew Kouri (GitHub handle: dpkouri or dpkouri@sandia.gov)
Denis Ridzal (GitHub handle: dridzal or dridzal@sandia.gov)

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

rol_python-2026.8.28.dev1935.tar.gz (3.9 MB view details)

Uploaded Aug 29, 2026 Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

rol_python-2026.8.28.dev1935-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (29.2 MB view details)

Uploaded Aug 29, 2026 CPython 3.14manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

rol_python-2026.8.28.dev1935-cp314-cp314-macosx_11_0_arm64.whl (13.2 MB view details)

Uploaded Aug 29, 2026 CPython 3.14macOS 11.0+ ARM64

rol_python-2026.8.28.dev1935-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (29.2 MB view details)

Uploaded Aug 29, 2026 CPython 3.13manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

rol_python-2026.8.28.dev1935-cp313-cp313-macosx_11_0_arm64.whl (13.2 MB view details)

Uploaded Aug 29, 2026 CPython 3.13macOS 11.0+ ARM64

rol_python-2026.8.28.dev1935-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (29.2 MB view details)

Uploaded Aug 29, 2026 CPython 3.12manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

rol_python-2026.8.28.dev1935-cp312-cp312-macosx_11_0_arm64.whl (13.2 MB view details)

Uploaded Aug 29, 2026 CPython 3.12macOS 11.0+ ARM64

rol_python-2026.8.28.dev1935-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (29.1 MB view details)

Uploaded Aug 29, 2026 CPython 3.11manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

rol_python-2026.8.28.dev1935-cp311-cp311-macosx_11_0_arm64.whl (13.1 MB view details)

Uploaded Aug 29, 2026 CPython 3.11macOS 11.0+ ARM64

rol_python-2026.8.28.dev1935-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (29.1 MB view details)

Uploaded Aug 29, 2026 CPython 3.10manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

rol_python-2026.8.28.dev1935-cp310-cp310-macosx_11_0_arm64.whl (13.1 MB view details)

Uploaded Aug 29, 2026 CPython 3.10macOS 11.0+ ARM64

rol_python-2026.8.28.dev1935-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (29.1 MB view details)

Uploaded Aug 29, 2026 CPython 3.9manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

rol_python-2026.8.28.dev1935-cp39-cp39-macosx_11_0_arm64.whl (13.1 MB view details)

Uploaded Aug 29, 2026 CPython 3.9macOS 11.0+ ARM64

rol_python-2026.8.28.dev1935-cp38-cp38-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (29.1 MB view details)

Uploaded Aug 29, 2026 CPython 3.8manylinux: glibc 2.27+ x86-64manylinux: glibc 2.28+ x86-64

rol_python-2026.8.28.dev1935-cp38-cp38-macosx_11_0_arm64.whl (13.1 MB view details)

Uploaded Aug 29, 2026 CPython 3.8macOS 11.0+ ARM64

File details

Details for the file rol_python-2026.8.28.dev1935.tar.gz.

File metadata

Download URL: rol_python-2026.8.28.dev1935.tar.gz
Upload date: Aug 29, 2026
Size: 3.9 MB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for rol_python-2026.8.28.dev1935.tar.gz
Algorithm	Hash digest
SHA256	`7e46dce40911dd14a6a642460b542541cc7e264013482dc12a22f54ec608b352`
MD5	`6d0ee75db9974bcbd852769bc6c7b991`
BLAKE2b-256	`9fb6cfb6c0c0c1f02e838cd7333868f37cef0dc9515a2907bd3d9c2c2a312e45`

Algorithm	Hash digest
SHA256	`ef979c08a2d8709cdec821468aaeedf0117318857481dd0303341b1d8081b86b`
MD5	`cba205ca4bc7c9f1a57c5d5d1efffe5a`
BLAKE2b-256	`96377b0fe34e80bc7483f6092161b5ae0e9c4c6109b4a0dd7bd252c8bdd70b49`

Algorithm	Hash digest
SHA256	`9c0ba2709792e8094fb7a14d9a10985c184223f0f52b026010d993cc3627c36b`
MD5	`129953f1bc63b8e3fafa7dc665763077`
BLAKE2b-256	`dc0e915f2676b486eb5025385361e599921692616bb9771a4264a2a797602373`

Algorithm	Hash digest
SHA256	`a7c42890c56e72493917c0910a934ca4fbf7fb2bff7d2e42e068bc7d2cd43a0a`
MD5	`a3762da23d8e6bff795bce0ced8df783`
BLAKE2b-256	`364eeb45014c8cf8f3aa733b8778e970f94c7eb03b00fa9c484df537579f04ed`

Algorithm	Hash digest
SHA256	`c62de60eeb3482df26656f8fb7db24e2c3002021893d20e887ad26363f295c95`
MD5	`dc54652d35c844afc40c41f2b8734f68`
BLAKE2b-256	`d711bac124d049a478b8cd471c87a7d26a995b098014d93d1b19c1e5797c704c`

Algorithm	Hash digest
SHA256	`e001d9afd882993b6ed765fe55ee8cd443afacaaf6f6a6cafc4c2d675b2e76c3`
MD5	`6c1216f994291026e26c89734137f74b`
BLAKE2b-256	`4df4383ab88021ca066cf5c78b23f4d7fc37186969a7d0f2bdc89d479f6d40ee`

Algorithm	Hash digest
SHA256	`18e206b37ad8ce26b6247fbd7da42c2787374cf804abe36b9fac8e976fc8c3a3`
MD5	`8b93a9cfc4362621b9b580cd55331465`
BLAKE2b-256	`2b811a15caae7d7b110e4706da553cd2f65fc4595c88826a5501cab9ba179a60`

Algorithm	Hash digest
SHA256	`8da7e660a297d930ae2b635ce0be399c88964394b149cfda2b24db3b5a410c9d`
MD5	`b79c9e91d6cce35acf0a4f932578a74b`
BLAKE2b-256	`027241458b7eb856dab9e8b0b33e0a3df4b807d4b3d0798364247dc5be6892a9`

Algorithm	Hash digest
SHA256	`7828815d89a1800ddf9cbb8af9534ca18b0624d9ff46f11167567d5d094d444d`
MD5	`659610af6183164d8d4be434db70e321`
BLAKE2b-256	`b0f2ff2a52ec462a8162864af90dca42f9a6f54f27defe66179470d8eb0fd108`

Algorithm	Hash digest
SHA256	`f094460fd6c0adfd1ca3a941fb2ece370a99fb36ef144916ebf63b0b605bccd9`
MD5	`d828388262c2f8bb4b97db4e34bbc9aa`
BLAKE2b-256	`31a80191fc9efa7b0470ee66a297ebc4ea5693f423d86f1424310f0480a86ebb`

Algorithm	Hash digest
SHA256	`3879e0f24d349288957198d0c12752835d3f1ef103cf9f61c0f4faee54a0ac76`
MD5	`b51f7b6557a393390189c2078342c385`
BLAKE2b-256	`531b91679919f93e1a879bab1b9cbc048f2101f1687721c1737571554a39ab7b`

Algorithm	Hash digest
SHA256	`ebf3fa613e0ef5aee44952030bfe6088ec0a8de8ad480924c2e10a5a97ee2589`
MD5	`02b2ba2b347cecb88231b8cfaa76946e`
BLAKE2b-256	`c9f184f0e7006ca0baf5da5ee46dd3ef9678ca1d7255f4b228ab54c7bfc00a65`

Algorithm	Hash digest
SHA256	`903473e48485c16eb4016328d55291f8d474dda341337d54cc41e7f52b1b3b03`
MD5	`daab2cd8cd71cb81605ec44e760dde74`
BLAKE2b-256	`3aa5d9db92811d0d43a92706bf647344333779b904254893eb1b4d3f2a8b3011`

Algorithm	Hash digest
SHA256	`02325bbdde0049a9e93f859d66967584da6f8f116f3a923c40f04f8dd2a120c8`
MD5	`deb6e9f29a637effdae924e5fd0464cb`
BLAKE2b-256	`96edea810d3668412271f608e18db28d45c1c9a40a3e64a0649dc329c9f0dac0`

Algorithm	Hash digest
SHA256	`6fee504fa81a4deefbd11e2a60d0ad2bbda5da68daa5f8cd3a38282510c03d5f`
MD5	`b5597f6306e0bf184e37cf3f0b783da0`
BLAKE2b-256	`49ce082ef384701219130fdc139b0903d0e146f9eb9e6829362829f34d47fd2c`

rol-python 2026.8.28.dev1935

Rapid Optimization Library (ROL)

Getting Started

Building against ROL

Python Interface

Copyright and License

Questions?

Key dates

2 maintainers

Credits

License expression

Requires

Classifiers

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

Release history Release notifications | RSS feed