This is the official implementation of the paper "FLUID-LLM: Learning Computational Fluid Dynamics with Spatiotemporal-aware Large Language Models".

@misc{zhu2024fluidllmlearningcomputationalfluid,
      title={FLUID-LLM: Learning Computational Fluid Dynamics with Spatiotemporal-aware Large Language Models}, 
      author={Max Zhu and Adrián Bazaga and Pietro Liò},
      year={2024},
      eprint={2406.04501},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2406.04501}, 
}

The recommended way to install the dependencies is to use the provided environment.yml file. To create a new conda environment, run create_conda_env.sh.

The two main branches are 'main' for training on the CFD datasets, and 'synth_ds' for training synthetic datasets.

For feedback, questions, or press inquiries please contact Max Zhu or Adrián Bazaga