Upload README.md with huggingface_hub

README.md

@@ -24,12 +24,12 @@ arxiv: 2601.09211
 
 ## Pretrained checkpoints
 
-| Model | Description | Size | Path |
-|-------|-------------|------|------|
-| `reconstruction` | Multi-view RGBD sparse-structure flow (DINOv2-ViT-L + DiT-B, 16³ voxel grid) | 628 MB | [`reconstruction/`](./tree/main/reconstruction) |
-| `affordance` | Text-conditioned affordance heatmap flow (CLIP-ViT-L/14 + DiT-B, 64³ sparse grid) | 707 MB | [`affordance/`](./tree/main/affordance) |
+Hosted at [`chrockey/Affostruction`](https://huggingface.co/chrockey/Affostruction). Each subfolder contains a flat `config.json` + `model.safetensors`. SLAT + mesh/gaussian decoders are pulled from [`microsoft/TRELLIS-image-large`](https://huggingface.co/microsoft/TRELLIS-image-large).
 
-Both subfolders contain a flat `config.json` + `model.safetensors`. SLAT + mesh/gaussian decoders are pulled from [`microsoft/TRELLIS-image-large`](https://huggingface.co/microsoft/TRELLIS-image-large).
+| Model | Description | Size |
+|-------|-------------|------|
+| `reconstruction` | Multi-view RGBD sparse-structure flow | 628 MB |
+| `affordance` | Text-conditioned affordance heatmap flow | 707 MB |
 
 ## Usage