InfoBay.AI – 3D CAD Dataset Catalogue
Overview
The InfoBay.AI – 3D CAD Dataset Catalogue is a professionally curated collection of CAD models covering engineering, manufacturing, industrial design, architecture, product development, and simulation. The dataset is intended for research, AI model training, digital twin applications, robotics, computer vision, and generative design.
This release contains representative samples from the complete enterprise-scale InfoBay.AI CAD collection.
Dataset Statistics
| Property | Value |
|---|---|
| Asset Type | 3D CAD Models |
| Total Assets | 1,200,362 |
| Storage Format | Apache Parquet |
| Compression | ZSTD |
| License | CC-BY-4.0 |
| Folder Hierarchy | Preserved |
| Metadata Included | Yes |
Categories
The dataset contains CAD assets from multiple engineering and industrial domains including:
- Mechanical Engineering
- Architectural Engineering
- Electrical Engineering
- Industrial Equipment
- Manufacturing Components
- Robotics
- Automotive Parts
- Aerospace Components
- Medical Equipment
- Consumer Products
- Characters & Creatures
- Art & Abstract
- Cultural Heritage
- Industrial Assemblies
Sample Assets
Examples include:
- Aircraft Landing Gear
- Heat Exchanger
- Pump
- Vehicle Suspension System
- Automatic Sliding Gate System
- Camera Lens
- Warehouse Utility Cart
- Steam Engine
- Aircraft Engine
- Injection Mold
- Boring Machine
- Coil Winding Machine
Parquet Schema
Each row represents one original file.
| Column | Type | Description |
|---|---|---|
| file_name | string | Original filename |
| relative_path | string | Original folder path |
| extension | string | File extension |
| file_size | int64 | File size in bytes |
| sha256 | string | SHA-256 checksum |
| content | binary | Original file content |
The original folder hierarchy is preserved through the relative_path column.
Dataset Features
- Large-scale industrial CAD collection
- Original folder structure retained
- Binary files stored losslessly
- Metadata searchable
- SHA-256 integrity verification
- Efficient Apache Parquet storage
- ZSTD compression
- Suitable for enterprise AI workflows
Applications
This dataset can be used for:
- CAD Retrieval
- Generative AI
- Digital Twin
- Engineering AI
- Robotics
- Manufacturing Automation
- Industrial Inspection
- Reverse Engineering
- 3D Shape Analysis
- Product Design Automation
- Simulation
- Spatial Computing
- AR / VR
- Game Development
Industries
- Manufacturing
- Automotive
- Aerospace
- Construction
- Industrial Automation
- Medical Devices
- Electronics
- Heavy Engineering
- Architecture
- Education
- Research
File Formats
The collection may contain formats including:
- OBJ
- GLB
- GLTF
- STL
- STEP
- STP
- IGES
- IGS
- FBX
- DAE
- PLY
- DXF
- DWG
- 3DS
Loading the Dataset
from datasets import load_dataset
dataset = load_dataset(
"YOUR_USERNAME/3d-cad-dataset",
split="train"
)
print(dataset)
Restoring the Original Dataset
Each row stores the original binary content together with its original relative path. The dataset can therefore be reconstructed exactly as it existed before conversion by writing the content field back to the corresponding relative_path.
Intended AI Use Cases
- Foundation Models for 3D Assets
- CAD Retrieval Systems
- Design Similarity Search
- Engineering Knowledge Graphs
- Large Multimodal Models
- Digital Twin Platforms
- Geometry Processing
- Robotics Simulation
- Industrial AI
License
This dataset is distributed under the Creative Commons Attribution 4.0 International (CC BY 4.0) License.
Users are free to:
- Share
- Adapt
- Redistribute
- Build commercial products
provided proper attribution is given.
Citation
@dataset{infobay_3d_cad_dataset_2026,
title = {InfoBay.AI – 3D CAD Dataset Catalogue},
author = {InfoBay.AI},
year = {2026},
license = {CC-BY-4.0},
publisher = {InfoBay.AI}
}
About InfoBay.AI
InfoBay.AI develops large-scale, enterprise-grade AI datasets spanning vision, language, speech, documents, healthcare, engineering, and multimodal applications. The datasets are designed to support research, foundation model training, enterprise AI solutions, and next-generation generative AI systems.
- Downloads last month
- -