FD-Bench: A Modular and Fair Benchmark for Data-driven Fluid Simulation

Data-driven modeling of fluid dynamics has advanced rapidly with neural PDE solvers, yet a fair and strong benchmark remains fragmented due to the absence of unified PDE datasets and standardized evaluation protocols. Although architectural innovations are abundant, fair assessment is further impeded by the lack of a clear disentanglement between spatial, temporal and loss modules. In this paper, we introduce FD-Bench, a large-scale, modular benchmarking infrastructure that unifies datasets, evaluation protocols, and reproducible pipelines for data-driven fluid simulation. FD-Bench systematically reviews and decomposes mainstream baseline models reported across recent publications, extracting and standardizing their key architectural and training components for fair, unified comparison across 10 representative flow scenarios. It provides four key contributions: (1) a modular design that enables fair comparisons across spatial, temporal, and loss function modules; (2) the first systematic framework for direct comparison with traditional numerical solvers; (3) fine-grained generalization analysis across resolutions, initial conditions, and prediction time window; and (4) a user-friendly, extensible codebase to support future research. Through rigorous empirical studies, FD-Bench establishes the most comprehensive leaderboard to date, resolving long-standing issues in reproducibility and comparability, and laying a foundation for robust evaluation of future data-driven fluid models. The code is open-sourced at https://github.com/WillDreamer/FD-Bench.