README.md

---
license: other
license_name: nvidia-open-model-license
license_link: >-
  https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-agreement/
---

# FIGConvNet DrivAerML Surface

FIGConvNet DrivAerML Surface is a deep learning model for predicting surface
aerodynamic fields on automotive geometries. The model predicts pressure and
wall shear stress fields on 3D vehicle surface meshes for computational fluid
dynamics (CFD) applications.

This model is available for commercial use.

### License/Terms of Use:

Use of this model is governed by the [NVIDIA Open Model Agreement](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-agreement/).

### Deployment Geography:

Global

### Use Case:

Computational Fluid Dynamics (CFD) engineers accelerating
automotive external aerodynamics with AI.

### Release Date:

05/01/2026

Hugging Face: https://huggingface.co/nvidia/figconvnet_drivaerml_surface

## Reference(s):

[Code](https://github.com/NVIDIA/physicsnemo/tree/main/examples/cfd/external_aerodynamics/figconvnet)

[Factorized Implicit Global Convolution for Automotive Computational Fluid
Dynamics Prediction](https://arxiv.org/abs/2502.04317)

[DrivAerML: High-Fidelity Computational Fluid Dynamics Dataset for Road-Car
External Aerodynamics](https://arxiv.org/abs/2408.11969)

## Model Architecture:

**Architecture Type:** FIGConvNet uses a U-Net architecture with factorized
implicit global convolutional layers.

**Network Architecture:** FIGConvUNet

**Number of model parameters:** 6,577,413



## Input:

**Input Type(s):**

- Tensor (3D point cloud coordinates on vehicle surface)

**Input Format(s):** PyTorch Tensor

**Input Parameters:**

- Three Dimensional (3D) (batch, num_points, 3)

**Other Properties Related to Input:**

- Input point cloud represents vehicle surface geometry with coordinates
  normalized to the bounding box: x ∈ [-2.0, 2.0], y ∈ [-1.8, 1.8],
  z ∈ [-1.5, 2.6]
- Typical input size: 500,000 points per vehicle geometry

## Output:

**Output Type(s):** Tensor (Surface aerodynamic fields)

**Output Format:** PyTorch Tensor

**Output Parameters:** Three Dimensional (3D) (batch, num_points, 4)

**Other Properties Related to Output:**

- Output channels: 1 pressure field + 3 wall shear stress components (x, y, z)
- Predictions correspond to time-averaged CFD simulation results

Our AI models are designed and/or optimized to run on NVIDIA
GPU-accelerated systems. By leveraging NVIDIA’s hardware (e.g. GPU cores)
and software frameworks (e.g., CUDA libraries), the model achieves faster
training and inference times compared to CPU-only solutions.

## Software Integration

**Runtime Engine(s):** PyTorch

**Supported Hardware Microarchitecture Compatibility:**

- NVIDIA Ampere
- NVIDIA Blackwell
- NVIDIA Hopper
- NVIDIA Turing

**Supported Operating System(s):**

- Linux

## Model Version(s):

**Model Version:** 1.0.0

# Training, Testing, and Evaluation Datasets:

The DrivAerML dataset is used for training and evaluation, which is a publicly available,
high-fidelity dataset comprising aerodynamic data for 500 parametrically morphed variants
of the DrivAer notchback vehicle. The dataset was generated using hybrid RANSLES (HRLES),
a scale-resolving CFD method, which provides time-averaged quantities for each variant.
The available data includes surface pressure, wall shear stress, and flow-field quantities,
provided in formats compatible with mesh-based analysis (.vtp for surface data and .vtu
for flow-field data). 10% of the samples are used as the test set, with 20% of the test set
consisting of out-of-distribution samples based on drag coefficients. These samples
represent extreme cases with the lowest and highest drag coefficients in the entire
dataset, which remain unseen by the model during training.

## Training Dataset:

**Data Modality:**

- Other: 3D Point Cloud

**Training Data Size:**

- 436 files in VTP format that contain meshes and corresponding physical quantities

**Link:** [DrivAerML Dataset](https://arxiv.org/abs/2408.11969)

**Data Collection Method by dataset:**

- Synthetic CFD Simulation

**Labeling Method by dataset:**

- Automated

**Properties:**
The data is a simulation/synthetic dataset generated using the
[OpenFOAM CFD solver](https://www.openfoam.com/news/main-news/openfoam-v2206)
to generate flow fields such as velocity and pressure for different car geometries
for the same boundary condition configuration as used to generate the training set.

## Evaluation Dataset:

**Link:** [DrivAerML Dataset](https://arxiv.org/abs/2408.11969)

**Data Collection Method by dataset:**

- Synthetic CFD Simulation

**Labeling Method by dataset:**

- Automated

**Properties:**
Validation split from DrivAerML dataset with vehicle geometries held out from
training. The full DrivAerML dataset is split as 90% for training
and 10% for validation.

## Inference:

**Engine:** PyTorch

**Test Hardware:**

- A100
- H100

## Ethical Considerations:

NVIDIA believes Trustworthy AI is a shared responsibility and we have
established policies and practices to enable development for a wide array of AI
applications.  When downloaded or used in accordance with our terms of service,
developers should work with their internal model team to ensure this model
meets requirements for the relevant industry and use case and addresses
unforeseen product misuse.

For more detailed information on ethical considerations for this model,
please see the Model Card++ subcards: [Bias](https://huggingface.co/nvidia/figconvnet_drivaerml_surface/blob/main/bias.md), [Explainability](https://huggingface.co/nvidia/figconvnet_drivaerml_surface/blob/main/explainability.md), [Privacy](https://huggingface.co/nvidia/figconvnet_drivaerml_surface/blob/main/privacy.md), and [Safety & Security](https://huggingface.co/nvidia/figconvnet_drivaerml_surface/blob/main/safety.md).

Please report security vulnerabilities or NVIDIA AI Concerns
[here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/).