datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
closure-challenge-v2-cfd-cases
Closure Challenge v2 — CFD case files (heavy)
Companion dataset to anon-closure-challenge-v2/closure-challenge-v2.
This repository hosts the full OpenFOAM cases (mesh, fields, run scripts), VTK volume snapshots, and DNS-native partitioned VTUs for the 14 test cases of the Closure Challenge v2 benchmark, together with the standardized training and validation sets under data/train/ and data/validation/.
The lightweight integral-profile portion needed for reproducing the scoring… See the full description on the dataset page: https://huggingface.co/datasets/anon-closure-challenge-v2/closure-challenge-v2-cfd-cases.leaf1-pretraining-cfdna
LEAF-1 pretraining corpus — cfDNA fragments (hg38)
This dataset contains the cell-free DNA (cfDNA) portion of the pretraining
corpus of LEAF-1. The sample manifest and checksums of this dataset are
also archived on Zenodo (10.5281/zenodo.23195009).
The other portions of the pretraining corpus: scATAC-seq on Zenodo
(10.5281/zenodo.23176597); TCGA
ATAC-seq is controlled access, its manifest and processing code are on Zenodo
(10.5281/zenodo.23195360).
2,301 plasma cfDNA samples… See the full description on the dataset page: https://huggingface.co/datasets/jing-z/leaf1-pretraining-cfdna.FuXi-CFD-dataset
FuXi-CFD Dataset
Overview
This dataset accompanies the paper:
Reconstructing fine-scale 3D wind fields with terrain-informed machine learning
It contains large-scale CFD-generated wind field simulations used for training and evaluating the FuXi-CFD model described in the paper.
The dataset includes over 12,000 cases.
Dataset Organization
The dataset is distributed as compressed archive files:… See the full description on the dataset page: https://huggingface.co/datasets/linchensen/FuXi-CFD-dataset.Bear-CFD-dataset
BEAR-CFD-Dataset
This dataset is released with paper Data-driven operator learning for energy-efficient and indoor air quality-aware building ventilation control.
The data was generated using ANSYS FLUENT 2023R2 and includes both steady-state and transient flow simulations. It is designed to support research in scientific machine learning, particularly in learning neural operators and data-driven PDE solvers.
This dataset is released for research purposes only. Please cite the… See the full description on the dataset page: https://huggingface.co/datasets/alwaysbyx/Bear-CFD-dataset.PhysicsNeMo-Datacenter-CFD
Overview
This asset provides the dataset and reference OpenFOAM configuration to be used for the Datacenter CFD use case here.
The dataset contains normalized OpenFOAM simulations for various configurations of a typical hot aisle in a datacenter. The dataset includes variation for the number of IT racks and hot-aisle geometry parameters like length, height and width of the hot aisle. These configurations are solved with OpenFOAM assuming maximum flow rate and rack exit temperature… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/PhysicsNeMo-Datacenter-CFD.ColdPlate-CFD-2000sims
Cold Plate CFD: 2000 Simulations
2,000 steady-state CFD simulations of a liquid-cooled, dual-chip, pin-fin cold
plate, plus the POD-NN surrogate trained on them and the code that reproduces
every figure in the paper.
Dataset DOI: 10.5061/dryad.k0p2ngfp5
· Primary article: AI Thermal Fluids, 10.1016/j.aitf.2026.100040
· Model: UARK-NED3/PODNN-ColdPlate
· License: CC0 1.0 (public domain dedication)
Mirror of the Dryad deposit published 29 April 2026.
What was simulated… See the full description on the dataset page: https://huggingface.co/datasets/UARK-NED3/ColdPlate-CFD-2000sims.ubend-cfd
U-bend Conjugate Heat Transfer Dataset
Parameterized U-bend CFD dataset for surrogate modeling and scientific machine learning.
Five randomly selected samples showing the geometry and the solution fields (velocity magnitude, pressure, temperature, turbulent kinetic energy, turbulent viscosity).
Dataset Description
This dataset contains ~9,000 conjugate heat transfer simulations of a parameterized U-bend cooling channel, generated with OpenFOAM… See the full description on the dataset page: https://huggingface.co/datasets/JDecke/ubend-cfd.CFDBench-raw
[!WARNING]
Parts of the data (the Cylinder flow) is still uploading.
Links:
Paper (arXiv)
GitHub
Interpolated data
This dataset contains the raw (before interpolation) data of the CFDBench dataset, from the paper CFDBench: A Large-Scale Benchmark for Machine Learning Methods in Fluid Dynamics, which is used for training and evaluating data-driven neural operators for fluid dynamics. This dataset is very large (approx. 460GB) and your likely only need the interpolated data if you just want… See the full description on the dataset page: https://huggingface.co/datasets/chen-yingfa/CFDBench-raw.260204_cfd_backup
CleanFD Backup
CleanFD is a comprehensive modularized library for benchmarking AI-generated image detection methods. This backup includes the complete project with all detectors, datasets, pretrained models, and evaluation scripts.
Download & Extract
# huggingface_hub 설치 (필요시)
pip install huggingface_hub
# 다운로드
huggingface-cli download leekwoon/260204_cfd_backup --repo-type dataset --local-dir ./cfd_backup
# 무결성 확인 (선택사항)
cd cleanfd_backup
md5sum -c checksums.md5
# 파일… See the full description on the dataset page: https://huggingface.co/datasets/leekwoon/260204_cfd_backup.CFDBenchLinks:
Paper
GitHub
Raw data
This dataset contains the interpolated data of the CFDBench dataset, from the paper CFDBench: A Large-Scale Benchmark for Machine Learning Methods in Fluid Dynamics, which is used for training and evaluating data-driven neural operators for fluid dynamics.
Citation
@article{CFDBench,
title={CFDBench: A Large-Scale Benchmark for Machine Learning Methods in Fluid Dynamics},
author={Yining, Luo and Yingfa, Chen and Zhen, Zhang}… See the full description on the dataset page: https://huggingface.co/datasets/chen-yingfa/CFDBench.cfd_benchmark
CFD Benchmark
Dataset Description
CFD Benchmark is a comprehensive benchmark dataset for training and evaluating neural PDE solvers. It contains six standard tasks on regular grids, structured grids, and irregular geometries. The dataset was used for unified evaluation in the ICML 2024 paper Transolver, and some of its data originates from the FNO and Geo-FNO works.
The dataset contains six subsets: airfoil, darcy, elasticity, ns, pipe, and plas.
Paper: Transolver:… See the full description on the dataset page: https://huggingface.co/datasets/OneScience-Group/cfd_benchmark.PhysicsNeMo-CFD-Ahmed-BodyAhmed body surface data includes 3D surface geometry, pressure and wall shear stress data for variations in the Ahmed body geometry and inlet Reynolds number. This dataset is a great choice because it is relatively small in size, yet provides valuable information about aerodynamic simulations. It's ideal for teaching and experimentation, as it won't demand excessive storage or computational power. This dataset is for R&D and demonstration purposes and not meant for production.
TransportBenchcfdbench
CFDBench
Dataset Description
CFDBench is a large-scale benchmark dataset for machine learning methods in computational fluid dynamics, designed to evaluate the generalization capabilities of neural operators under unseen boundary conditions, fluid properties, and geometries.
The dataset contains four classic CFD problems: lid-driven cavity flow (cavity), laminar pipe flow (tube), step dam-break flow (dam), and flow around a cylinder (cylinder). For each problem… See the full description on the dataset page: https://huggingface.co/datasets/OneScience-Group/cfdbench.CF-Div2-Stepfun
CF-Div2-Stepfun Evaluation Benchmark
Offline benchmark of 53 Div.2 CodeForces problems.
Introduction
We introduce CF-Div2-Stepfun, a dataset curated to benchmark the competitive programming capabilities of Large Language Models (LLMs). We evaluate our proprietary Step 3.5 Flash alongside several frontier models on this benchmark.
The benchmark comprises 53 problems sourced from official CodeForces Division 2 contests held between September 2024 and February 2025. We… See the full description on the dataset page: https://huggingface.co/datasets/stepfun-ai/CF-Div2-Stepfun.cfduankouwind-cfd-trial
Wind Flow Over Complex Terrain Dataset
A large-scale dataset of steady-state RANS wind flow simulations over real-world complex terrain, designed for training machine learning models for wind resource assessment and atmospheric flow prediction.
Overview
Parameter
Value
Number of terrain locations
~1000
Wind directions per terrain
2 (random)
Total simulation cases
~10,000
Cropped grid per case
~298 × 298 × 64
Horizontal resolution (AOI)
~30 m
Vertical… See the full description on the dataset page: https://huggingface.co/datasets/souravsud/wind-cfd-trial.cfd-chaotic-advection
Chirikov Standard Map Lyapunov Spectrum
Maximal Lyapunov exponent Λ(K) for the Chirikov standard map on T², computed with a custom CUDA Benettin kernel on NVIDIA RTX 5090 (sm_120).
Part of the bigcompute.science CFD conjecture program — GPU-accelerated exploration of open questions in fluid mixing and dynamical systems.
Quick Start
from datasets import load_dataset
ds = load_dataset("cahlen/cfd-chaotic-advection", "deep_sweep", split="train")
row = ds[1200]… See the full description on the dataset page: https://huggingface.co/datasets/cahlen/cfd-chaotic-advection.Adult_train_cfda_instbasedlm_66
Dataset Card for "Adult_train_cfda_instbasedlm_66"
More Information needed
CFD
Navier-Stokes Simulated Flow Dataset for PINNs
Welcome to the Dataset!
Dive into the dynamic world of fluid flow with the Navier-Stokes Simulated Flow Dataset for PINNs! This collection of 10,000 simulated data points captures the essence of fluid dynamics in a 2D channel, tailored specifically for training Physics-Informed Neural Networks (PINNs). With an even split of 5,000 laminar flow and 5,000 turbulent flow points, this dataset is perfect for researchers, data… See the full description on the dataset page: https://huggingface.co/datasets/AethronPhantom/CFD.CFD_Dataset
Dataset Card for "CFD_Dataset"
More Information needed
cfd-ns3d-bkm
3D Navier–Stokes BKM Blowup Search
Three-dimensional vorticity-form incompressible Navier–Stokes with vortex stretching (ω⋅∇)u(\boldsymbol{\omega}\cdot\nabla)\mathbf{u}(ω⋅∇)u, pseudospectral cuFFT C2C on a periodic cube, and BKM integral tracking. Custom CUDA kernel on NVIDIA RTX 5090.
∂tω+(u⋅∇)ω=(ω⋅∇)u+ν∇2ω
\partial_t \omega + (\mathbf{u}\cdot\nabla)\omega = (\boldsymbol{\omega}\cdot\nabla)\mathbf{u} + \nu \nabla^2 \omega
∂tω+(u⋅∇)ω=(ω⋅∇)u+ν∇2ω
u^=i(k×ω^)∥k∥2
\hat{\mathbf{u}}… See the full description on the dataset page: https://huggingface.co/datasets/cahlen/cfd-ns3d-bkm.cf_data_2curly-fm-cfdFuXi-CFD-dataset
FuXi-CFD Dataset
Overview
This dataset accompanies the paper:
Reconstructing fine-scale 3D wind fields with terrain-informed machine learning
It contains large-scale CFD-generated wind field simulations used for training and evaluating the FuXi-CFD model described in the paper.
The dataset includes over 12,000 cases.
Dataset Organization
The dataset is distributed as compressed archive files:… See the full description on the dataset page: https://huggingface.co/datasets/ShuoLi-CityU/FuXi-CFD-dataset.cfd-ns-bkm
2D Navier–Stokes BKM Diagnostic
Pseudospectral vorticity-form Navier–Stokes on a periodic torus [0,2π)2[0,2\pi)^2[0,2π)2, with the Beale–Kato–Majda (BKM) diagnostic
∫0T∥ω(⋅,t)∥L∞ dt
\int_0^T \lVert \omega(\cdot,t) \rVert_{L^\infty}\, dt
∫0T∥ω(⋅,t)∥L∞dt
tracked alongside enstrophy. Computed with a custom CUDA + cuFFT kernel on NVIDIA RTX 5090 (Blackwell, sm_120).
Part of the bigcompute.science CFD conjecture program — GPU infrastructure toward 3D BKM blowup searches. 2D… See the full description on the dataset page: https://huggingface.co/datasets/cahlen/cfd-ns-bkm.cfduankouafrica-morocco-debit-moyen-journalier-de-la-station-ait-melloul-1970-2024-cfdc3730
Debit Moyen Journalier De La Station Ait Melloul 1970 2024 | Africa (Morocco Open Data)
34 rows - 1 Africa country/area - time not specified - source table - Engineered by Electric Sheep Africa
TL;DR
This dataset contains 34 rows from Morocco Open Data, covering Debit Moyen Journalier De La Station Ait Melloul 1970 2024. It is published as ML-ready Parquet with consistent Hugging Face metadata, source provenance, and analysis-friendly loading examples.… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-morocco-debit-moyen-journalier-de-la-station-ait-melloul-1970-2024-cfdc3730.cfd-surroagte-datacf_data