SAT solvers from the first competitions to Knuth's programs, each as a Docker image with a binary ready to run, the most thorough traceability we can offer (original sources, dated build environment, recipe and verification), and reproducible in a few clicks on your own machine.

A collective site, still being completed: please check, correct and contribute. Everything on these pages is generated from the data of the sat-heritage/docker-images repository as it stands today: which solvers build, which answer correctly, who wrote them, under which licence, with which awards. Much of it was assembled recently, year by year, partly with the help of an AI assistant, and is still being verified. Treat every claim as provisional and check it against the original competition material before relying on it. We strongly encourage pull requests and verification reports: a corrected author line, a licence, a working recipe, a missing solver, or the outcome of running an image on your own machine are all welcome as pull requests or issues. The goal is a site maintained by the SAT community as a whole. Images not on Docker Hub yet can be built locally with satex build <solver>:<year> (pip install satex).
Catalogue › 2013

CSHCrandMC

Yuri Malitsky, Ashish Sabharwal, Horst Samulowitz, Meinolf Sellmann.
solverotherrandom1st SAT+UNSAT · random track 2013
licenceGPL-2.0GPL-3.0LGPL-2.1MIT
can doSAT/UNSAT (declared)gzip input
statusNot run yetnot buildableNot on Docker Hub yet

Awards

Status

FIXME

Notes

Launchcommand was: python .//algport.py --tmpdir <tempdir> CSHCrandMC <instance> Track was Core solvers, Sequential, Random SAT

Licence

copyleft (GPL) Components under different licences: the strictest one rules the whole. You can use and modify it freely, but a program you redistribute with this solver inside must be GPL too, with its sources.

Read from licence file CSHCrandMC-src/code/utilities/Satzilla_features/satfeatures/bison-2.3/COPYING, CSHCrandMC-src/code/CSHCrandMC/DATASTRUCTURES/license.txt, CSHCrandMC-src/code/utilities/Satzilla_features/satfeatures/SAT-features-competition2012/lp_solve_4.0/LICENSE, CSHCrandMC-src/code/utilities/Satzilla_features/satfeatures/SAT-features-competition2012/VARSAT/mtl/zlib-1.2.3/contrib/dotzlib/LICENSE_1_0.txt; this summary is informative, the licence text prevails. Corrections welcome by pull request.

Build it and run it

pip install satex && satex build cshcrandmc:2013
docker run --rm -v $PWD:/data satex/cshcrandmc:2013 instance.cnf

This image is not on Docker Hub yet (checked 2026-09-25): the images are pushed in batches, and some entries cannot be built. Until then, satex build makes it on your machine from the archived sources and the recipe below, and the run command is the same. Mount the directory that holds your instance on /data; the proof file is optional and not produced by this solver. Its full provenance is kept: the archived sources, the pinned build environment and the recipe are all listed below, and satex build cshcrandmc:2013 rebuilds the same image on your own machine if you would rather not trust ours (slower, same solver).

Download the sources 98-CSHCrandMC.zip, the competition submission as archived by SAT Heritage, to build it yourself with the recipe below. The Zenodo record holding the archives of this year was downloaded 5 086 times (Zenodo counts per record, not per file).

Licence
GPL-2.0, GPL-3.0, LGPL-2.1, MIT (licence file CSHCrandMC-src/code/utilities/Satzilla_features/satfeatures/bison-2.3/COPYING, CSHCrandMC-src/code/CSHCrandMC/DATASTRUCTURES/license.txt, CSHCrandMC-src/code/utilities/Satzilla_features/satfeatures/SAT-features-competition2012/lp_solve_4.0/LICENSE, CSHCrandMC-src/code/utilities/Satzilla_features/satfeatures/SAT-features-competition2012/VARSAT/mtl/zlib-1.2.3/contrib/dotzlib/LICENSE_1_0.txt)
Image
satex/cshcrandmc:2013
Command
python .//algport.py --tmpdir /tmp CSHCrandMC FILECNF
Compressed input
read natively

How it is built

Recipe
builder generic/v1
Builder
generic/v1
Environment
debian:8.6
Sources
98-CSHCrandMC.zip

Last verification

No recorded run.