Dataset · codepawl
Index of 37,688 3D Assets Across Six Libraries
codepawl/asset-index
A unified metadata table for searching and selecting 3D models by size, style, and characteristics.
What's inside
The dataset combines metadata for 37,688 3D assets from six libraries (including 22,696 models from Quixel Megascans) into a single table. Rows indicate real-world dimensions in meters, type, style, surface, biome, PBR maps, polygon count, rigging status, and file acquisition method. Modalities include tabular data, text, and 3D.
What you can do with it
- Train AI agents for automated 3D asset selection for scenes
- Build semantic search across 3D libraries for game dev and animation
- Size
- 10K–100K rows
- Downloads this month
- 2.1K
- Likes
- 27
How to load it
from datasets import load_dataset
ds = load_dataset("codepawl/asset-index")