A Status Quo Investigation of Large Language Models towards Cost-Effective CFD Automation with OpenFOAMGPT: ChatGPT vs. Qwen vs. Deepseek
This work addresses the problem of automating specialized CFD simulations for engineers and researchers, but it is incremental as it highlights current limitations rather than breakthroughs.
The study evaluated multiple large language models for automating CFD tasks with OpenFOAMGPT, finding that while some models efficiently handled tasks like adjusting boundary conditions and solver configurations, token costs and stability varied, and zero-shot prompting often failed for complex simulations requiring expert supervision.
We evaluated the performance of OpenFOAMGPT incorporating multiple large-language models. Some of the present models efficiently manage different CFD tasks such as adjusting boundary conditions, turbulence models, and solver configurations, although their token cost and stability vary. Locally deployed smaller models like QwQ-32B struggled with generating valid solver files for complex processes. Zero-shot prompting commonly failed in simulations with intricate settings, even for large models. Challenges with boundary conditions and solver keywords stress the requirement for expert supervision, indicating that further development is needed to fully automate specialized CFD simulations.