This repository contains the official implementation for the following three papers (you can use branches to access the other versions):
-
V0.1: OptiMUS: Optimization Modeling Using mip Solvers and large language models.
-
V0.2: OptiMUS: Scalable Optimization Modeling with (MI) LP Solvers and Large Language Models.
-
V0.3: OptiMUS-0.3: Using Large Language Models to Model and Solve Optimization Problems at Scale
Live demo: https://optimus-solver.com/
You can download the dataset from https://huggingface.co/datasets/udell-lab/NLP4LP. Please note that NLP4LP is intended and licensed for research use only. The dataset is CC BY NC 4.0 (allowing only non-commercial use) and models trained using the dataset should not be used outside of research purposes.
OptiMUS has two available implementations
OptiMUS v1 adopts a sequential work-flow implementation. Suitable for small and medium-sized problems.
@article{ahmaditeshnizi2023optimus,
title={OptiMUS: Optimization Modeling Using mip Solvers and large language models},
author={AhmadiTeshnizi, Ali and Gao, Wenzhi and Udell, Madeleine},
journal={arXiv preprint arXiv:2310.06116},
year={2023}
}
OptiMUS v2 adopts agent-based implementation. Suitable for large and complicated tasks.
@article{ahmaditeshnizi2024optimus,
title={OptiMUS: Scalable Optimization Modeling with (MI) LP Solvers and Large Language Models},
author={AhmadiTeshnizi, Ali and Gao, Wenzhi and Udell, Madeleine},
journal={arXiv preprint arXiv:2402.10172},
year={2024}
}
OptiMUS v3 adds RAG and large-scale optimization techniques.
@article{ahmaditeshnizi2024optimus,
title={OptiMUS-0.3: Using Large Language Models to Model and Solve Optimization Problems at Scale},
author={AhmadiTeshnizi, Ali and Gao, Wenzhi and Brunborg, Herman and Talaei, Shayan and Udell, Madeleine},
journal={arXiv preprint arXiv:2407.19633},
year={2024}
}