ArchCAD: A Multimodal Dataset for Vectorized Engineering Drawing Understanding

40,000 samples with five strictly aligned modalities provide foundational data for AI systems to interpret CAD drawings. The dataset, created by jackluoluo and last updated in October 2025, is designed to address the challenge of understanding and utilizing computer-aided design data.

Use Cases

• Train multimodal AI models based on the five aligned data modalities mentioned in the description
• Develop visual question answering systems for engineering drawings based on the dataset's structure
• Benchmark image-to-text generation models on technical CAD diagrams
• Research foundational AI understanding of vectorized drawings based on the dataset's stated purpose

Strengths

• 40,000 samples provide a substantial base for model training
• Five strictly aligned modalities offer a consistent multimodal foundation

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment

Provenance

Source

huggingface

Collection Method

Method of gathering is not described in the provided input.

Time Range

Temporal coverage is not specified in the provided input.

Freshness

Last updated 2025-10-17 14:57:00

Geography

Spatial coverage is not specified in the provided input.