Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?


Failed to load latest commit information.
Latest commit message
Commit time

REMIND - REgional Model of INvestments and Development

CodeCheck Status


The REgional Model of INvestments and Development (REMIND) is a modular open source multi-regional model incorporating the economy, the climate system and a detailed representation of the energy sector. It solves for an inter-temporal Pareto optimum in economic and energy investments in the model regions, fully accounting for interregional trade in goods, energy carriers and emissions allowances. REMIND allows for the analysis of technology options and policy proposals for climate mitigation. The macro-economic core of REMIND is a Ramsey-type optimal growth model in which intertemporal global welfare is optimized subject to equilibrium constraints.


The model documentation for version 3.0.0 can be found at .

The most recent version of the documentation can also be extracted from the model source code via the R package goxygen ( To extract the documentation, run make docs in the main folder of the model. The resulting documentation can be found in the folder doc/.

Please pay attention to the REMIND Coding Etiquette when you modify the code (if you plan on contributing to the REMIND official repository). The Coding Etiquette is found in the documentation section of the file main.gms. It explains also the used name conventions and other structural characteristics. To automatically check if some aspects of the coding etiquette, you can run make check in the main folder of the model.


Tutorials can be found in the form of markdown files in the repository:


Copyright 2006-2022 Potsdam Institute for Climate Impact Research (PIK)


This program is free software: you can redistribute it and/or modify it under the terms of the GNU Affero General Public License as published by the Free Software Foundation, version 3 of the License or later. You should have received a copy of the GNU Affero General Public License along with this program. See the LICENSE file in the root directory. If not, see

Under Section 7 of AGPL-3.0, you are granted additional permissions described in the REMIND License Exception, version 1.0 (see LICENSE file).


Following the principles of good scientific practice it is recommended to make the source code available in the events of model based publications or model-based consulting.

When using a modified version of REMIND which is not identical to versions in the official main repository at add a suffix to the name to allow distinguishing versions (format REMIND-suffix).


The full model is quite resource heavy and works best on machines with high CPU clock and memory. Recommended is a machine with Windows, MacOS or Linux, with at least 16GB of memory and a Core i7 CPU or similar.


Please refer to the installation guide.


Model run settings are set in config/default.cfg and main.gms (or another config file of the same structure). New model scenarios can be created by adding a row to config/scenario_config.csv


To run the model execute Rscript start.R (or source("start.R") from within R) in the main folder of the model. Make sure that the config file has been set correctly before starting the model.


We are interested in working with you! Contact us through GitHub ( or by E-mail ( if you have found and/or fixed a bug, developed a new model feature, have ideas for further model development or suggestions for improvements. We are open to any kind of contribution. Our aim is to develop an open, transparent and meaningful energy-economy-model, and to get a better understanding of the underlying processes and possible futures. Join us in doing so!


Model dependencies must be publicly available and should be Open Source. Development aim is to rather minimize than expand dependencies on non-free and/or non open source software. That means that besides currently existing dependencies on GAMS, the gdxrrw R package and the corresponding solvers there should be no additional dependencies of this kind and that these existing dependencies should be resolved in the future if possible.

If a new R package is added as dependency this package should fulfill the following requirements:

  • The package is published under an Open Source license
  • The package is distributed through CRAN or PIK-CRAN (the PIK-based, but publicly available package repository).
  • The package source code is available through a public, version controlled repository such as GitHub

For other dependencies comparable measures should apply. When a dependency is added this dependency should be added to the HOW TO INSTALL section in the README file of the model (mentioning the depencendy and explaining how it can be installed). If not all requirements can be fulfilled by the new dependency this case should be discussed with the model maintainer ( to find a good solution for it.


In order to allow other researchers to reproduce and use work done with REMIND one needs to make sure that all components necessary to perform a run can be shared. One of these components is the input data. As proprietary data usually does not allow its free distribution it should generally be avoided.

When adding a new data source, make sure that it can be freely shared with others. If this is not the case please consider using a different source or solution.

Data preparation should ideally be performed with the madrat data processing framework ( This makes sure that the processing is reproducible and links properly to the already existing data processing for REMIND.

In case that these recommendations can not be followed we would be happy if you could discuss that issue with the REMIND development team (


By default the results for a model run are written to an individual results folder within the "output/" folder of the model. The two most important output files are the fulldata.gdx and the REMIND_generic_scenario-name.mif. The fulldata.gdx is the technical output of the GAMS optimization and contains all quantities that were used during the optimization in unchanged form. The mif-file is a csv file of certain format and is synthetized from the fulldata.gdx by post-processing scripts. It can be read in any text editor or spreadsheet program and is well suited for a quick look at the results and for further analysis.




Please contact


See file CITATION.cff or the documentation of the model for information how to cite the model.



See list of authors in CITATION.cff


See log on GitHub (