3d-assets
3D_LLM_Diffusion-trimodal-assets-v11
3D LLM Diffusion: Tri-Modal Assets v11
This dataset repository contains the large data artifacts required to reproduce
the v11 text/XRD/crystal routing layer and train the released Qwen latent
adapter. It accompanies
Yangfan78/3D_LLM_Diffusion-trimodal-qwen-v11.
Contents
Path
Contents
embeddings/trimodal_cond_emb_v11/
final v11 train/validation/benchmark latent tables
corpus/mm_corpus/
27,136 train and 9,047 validation text records plus alignment… See the full description on the dataset page: https://huggingface.co/datasets/Yangfan78/3D_LLM_Diffusion-trimodal-assets-v11.MESHY.AI_800_GLB_3D-Assets_Categorised_and_Labelled
MESHY_GLB.zip - 809 Samples GLB/GLTF (Textured, Categorised).
MESHY_PLY.zip - 788 Samples PLY (Vertex Colored, Uncategorised).
This dataset is also available in vertex color projected PLY files with an open source model browser that makes the task of creating hand-picked subsets of this dataset fast and easy, you can download it here at: https://archive.org/details/meshy-collection-1.7z
I curated this selection from assets generated by members of the MESHY.AI Discord server.
I mostly… See the full description on the dataset page: https://huggingface.co/datasets/tfnn/MESHY.AI_800_GLB_3D-Assets_Categorised_and_Labelled.3d-demo-assets3d-1k-assets
3D-1K Assets
A curated 3D rigid-object asset library with a 10-category taxonomy and 1,000 GLB assets (one per base object class). Generated by a text → image → RGBA → 3D pipeline (FLUX.2-klein → RMBG-2.0 → TRELLIS.2).
This is the base class layer of a planned 1,000 × 1,000 = 1,000,000 instance dataset for embodied AI / robotic manipulation.
Files
file
size
content
glbs.tar
6.5 GB
1000 GLB assets: 0001.glb … 1000.glb
rgba_images.tar
700 MB
1000 RGBA… See the full description on the dataset page: https://huggingface.co/datasets/Dennis0626/3d-1k-assets.dream-network-prop-3d-assets
Dream Network — Isolated 3D Asset Prop Renders (20)
20 clean, isolated 3D prop and weapon reference renders from the Dream Network universe on pure neutral studio backgrounds. Designed specifically for Image-to-3D pipelines (Tripo3D, Trellis, InstantMesh, Hunyuan3D, Rodin, CSM, NeRF/Gaussian Splatting) and 3D modeling texture reference.
All cards, frames, text overlays, and background clutter have been stripped away, leaving only the centered object with crisp silhouettes, clean… See the full description on the dataset page: https://huggingface.co/datasets/TheMindExpansionNetwork/dream-network-prop-3d-assets.Blender_3D_assets