Evaluating LLMs for Translating Legacy Fortran Atmospheric Physics to GPU-Accelerated JAX

@misc{linck2026evaluating,
  title = {{Evaluating LLMs for Translating Legacy Fortran Atmospheric Physics to GPU-Accelerated JAX}},
  author = {Iris Linck and John Dennis and John Schreck and Karen Stengel and David John Gagne II},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.12662},
  eprint = {2610.12662},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.12662},
  note = {CC BY 4.0},
}

TY  - UNPB
TI  - Evaluating LLMs for Translating Legacy Fortran Atmospheric Physics to GPU-Accelerated JAX
AU  - Iris Linck
AU  - John Dennis
AU  - John Schreck
AU  - Karen Stengel
AU  - David John Gagne II
PY  - 2026
DA  - 2026/10/08
PB  - arXiv
DO  - 10.48550/arxiv.2610.12662
UR  - https://arxiv.org/abs/2610.12662
AB  - Legacy Fortran physics packages in atmospheric models such as CESM, WRF, and E3SM pose a barrier to GPU modernization, while manual porting requires accelerator-programming expertise. We present LAFT (LLM-Assisted Fortran Translation), an agentic framework for converting Fortran atmospheric physics parameterizations into GPU-accelerated JAX while preserving numerical agreement with the Fortran reference. LAFT coordinates source-code analysis, translation, repair, runtime validation, and performance profiling. It combines a deterministic front end for metadata extraction and bridge generation, structured translation prompts, numerical validation, and profiling-guided feedback to identify inefficient execution patterns. The bridge handles array-layout conversion separately from scientific computation. We evaluate LAFT using the Kessler warm-rain microphysics scheme and four LLMs: Claude, GPT, Gemini, and Qwen. All four validated translations reproduce the Fortran reference outputs to near floating-point precision for the evaluated case, but their GPU performance differs substantially with execution design. The Gemini-generated implementation achieves up to a 357-fold speedup over serial Fortran in cached execution at 1,000,000 atmospheric columns, excluding compilation and host-device transfers. The results demonstrate that numerical agreement alone does not ensure efficient GPU execution and motivate combining numerical validation with hardware-aware performance analysis. LAFT provides a practical workflow for modernizing legacy atmospheric physics code through LLM-assisted translation, deterministic interface generation, automated validation and repair, and reproducible profiling.
ER  -