Release weights, config and model card

README.md

@@ -0,0 +1,75 @@
+---
+license: other
+license_name: fair-noncommercial-research-license
+license_link: https://github.com/Jingong-Chen/GradSplat/blob/main/LICENSE
+library_name: gradsplat
+pipeline_tag: image-to-3d
+tags:
+  - gaussian-splatting
+  - novel-view-synthesis
+  - 3d-reconstruction
+  - feed-forward
+---
+
+# GradSplat
+
+**GradSplat** reconstructs a scene as 3D Gaussians from a few images in one forward pass, and refines its own prediction
+by learned gradient steps on the input views inside that pass, first on opacity and color and then on the shape of the
+Gaussians. No target view is used, so the steps run unchanged at inference.
+
+[Paper](https://arxiv.org/abs/2610.38951) · [Code](https://github.com/Jingong-Chen/GradSplat) · [Project page](https://jingong-chen.github.io/GradSplat/)
+
+These are the weights of the paper, trained on RealEstate10K. The DL3DV fine-tuned weights are at
+[Jingong/GradSplat-DL3DV](https://huggingface.co/Jingong/GradSplat-DL3DV).
+
+## Usage
+
+```bash
+git clone https://github.com/Jingong-Chen/GradSplat.git && cd GradSplat && pip install -e .
+```
+
+```python
+from gradsplat import GradSplat
+from gradsplat.io import load_views
+
+model = GradSplat.from_pretrained("Jingong/GradSplat").to("cuda").eval()
+views = load_views("examples/re10k_dining_room/images", cameras="examples/re10k_dining_room/cameras.json")
+out = model.infer(views.images, views.c2w, views.K)     # without cameras: model.infer(views.images)
+gaussians = out["gaussians"]                              # one Gaussian per input pixel
+```
+
+## Model
+
+- 1.22 B parameters: a VGGT-Ω backbone, a camera head, a dense head with a Gaussian branch, a shape head and two
+  learned updates (three appearance steps and five shape steps).
+- Input: 2 or more overlapping views of 224 × 224 pixels (center-cropped), with or without their cameras.
+- Output: one 3D Gaussian per input pixel (center, scale, rotation, opacity and spherical harmonics of degree 1), and
+  the predicted cameras.
+- Training: initialized from VGGT-Ω and trained for 178,000 steps in four stages on the training split of
+  RealEstate10K, with 2 to 12 input views per sample.
+
+## Results (from the paper)
+
+| | Narrow protocol (7,128 scenes) | Wide protocol (1,600 scenes) |
+|:---|:---:|:---:|
+| PSNR / SSIM / LPIPS, with input poses | 33.683 / 0.9593 / 0.0640 | 26.755 / 0.8993 / 0.1301 |
+| PSNR / SSIM / LPIPS, without input poses | 26.914 / 0.8863 / 0.1053 | 20.901 / 0.6945 / 0.2426 |
+
+RealEstate10K with six input and eight target views per scene at 224 × 224 pixels. Without fine-tuning: DL3DV
+24.787 / 0.8448 / 0.1919 and ScanNet++ 21.609 / 0.7540 / 0.3184.
+
+## License
+
+FAIR Noncommercial Research License, inherited from VGGT-Ω: noncommercial research use only.
+
+## Citation
+
+```bibtex
+@article{chen2026gradsplat,
+  title   = {GradSplat: Gradients Inside the Forward Pass for Feed-Forward 3D Gaussian Splatting},
+  author  = {Chen, Jingong and Jiang, Linlian and Li, Guangyuan and Zuo, Xinxin and
+             Wang, Yang and Pack, Chulwoo and Gao, Kyle and Won, Kwanghee},
+  journal = {arXiv preprint arXiv:2610.38951},
+  year    = {2026}
+}
+```

config.json

@@ -0,0 +1,23 @@
+{
+  "resolution": 224,
+  "patch_size": 16,
+  "embed_dim": 1024,
+  "sh_degree": 1,
+  "scale_max": 0.3,
+  "scale_mult": 0.001,
+  "dmean_scale": 0.05,
+  "gauss_upsample": 2,
+  "app_trunk_depth": 6,
+  "shape_hidden": 128,
+  "shape_amp": 4.0,
+  "shape_rot_amp": 1.0,
+  "appearance_steps": 3,
+  "shape_steps": 5,
+  "appearance_update_hidden": 192,
+  "shape_update_hidden": 128,
+  "mixer_depth": 4,
+  "mixer_dim": 768,
+  "mixer_heads": 12,
+  "mixer_mid": 64,
+  "kappa": 4.04
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2bbd05a8144504c0495fed1759d5a551f4db176d18ffa378168205dc2463b76
+size 4900169572