Solv3R: 3D Looped Transformer as a Multi-view Solver

Pretrained weights of Solv3R (paper, project page, code).

from solv3r import Solv3R, cameras_from_pose_enc, list_images, load_images

model = Solv3R.from_pretrained()                       # downloads solv3r.safetensors from this repo
images = load_images(list_images("path/to/images"))
ctx = model.encode(images)
z = model.solve(ctx, calls=16)
pred = model.decode(z, ctx)                            # depth, depth_conf, rays, world_points, pose_enc
c2w, K = cameras_from_pose_enc(pred["pose_enc"], (ctx.H, ctx.W))

The weights are built on DVLT and are released under the NVIDIA License (non-commercial, research and evaluation use only).

@article{kwak2026solv3r,
  title   = {Solv3R: 3D Looped Transformer as a Multi-view Solver},
  author  = {Kwak, Jeong-gi and Tagliasacchi, Andrea and Kagami, Sho and Ono, Yuki and Yi, Kwang Moo},
  journal = {arXiv preprint arXiv:2611.02857},
  year    = {2026}
}