WorldScore benchmark (ICCV 2025) · separate track
WorldScore metrics — world-generation quality on WorldScore's own camera tasks
WorldScore is a different benchmark from the PermaVid revisit set. It runs on its own scenes (8 static camera tasks: push-in / pull-out / move / orbit / pan) and scores an overall WorldScore plus 7 dimensions (no per-frame GT — these are learned/perceptual metrics). 9 of our methods were registered as generators and scored here.
⚠ Scope notes: (1) evaluated on 8 examples (one per camera layout), a subset of the full 2000-task leaderboard. (2) The VACE shown is the DL3DV long-horizon driver checkpoint — our UE/PermaVid finetune (ours_vace) was not run on WorldScore. (3) The dedicated finetuned run
vace14b_ft_lh@4000 returned all-zeros (eval did not complete) → excluded pending a rerun.overallWorldScore leaderboard (this subset)
metricsAll 7 dimensions per method
| # | method | WorldScore↑ | camera control | object control | content align | 3D consist. | photometric | style consist. | subjective | n |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | NVS-Solver 🏆 | 70.4 | 69.4 | 65.6 | 39.3 | 91.6 | 81.2 | 94.2 | 51.6 | 8 |
| 2 | VACE-14B (DL3DV-lh ckpt) | 67.8 | 78.3 | 56.2 | 33.0 | 85.4 | 88.9 | 93.2 | 39.8 | 8 |
| 3 | FlexWorld | 65.1 | 79.5 | 34.4 | 27.3 | 88.3 | 89.6 | 88.1 | 48.4 | 8 |
| 4 | Voyager | 63.3 | 75.0 | 50.0 | 33.9 | 80.5 | 83.5 | 89.2 | 31.1 | 8 |
| 5 | ViewExtrapolator | 61.1 | 76.6 | 59.4 | 32.7 | 79.5 | 56.0 | 87.0 | 36.7 | 8 |
| 6 | TrajectoryCrafter | 56.3 | 33.2 | 37.5 | 27.2 | 77.2 | 82.1 | 88.0 | 48.8 | 8 |
| 7 | GEN3C | 44.4 | 53.4 | 40.6 | 11.1 | 53.2 | 38.8 | 84.1 | 29.5 | 8 |
| 8 | FrameCrafter | 41.1 | 51.0 | 46.9 | 15.2 | 37.2 | 0.0 | 91.5 | 46.2 | 8 |
| 9 | ViewCrafter | 32.3 | 20.9 | 37.5 | 22.2 | 32.4 | 0.0 | 56.0 | 57.0 | 8 |
Cells shaded green (high) → red (low) per value (0–100). NVS-Solver tops this subset (70.4); VACE-14B (DL3DV-lh) is 2nd (67.8). These are perceptual / camera-control / consistency scores on WorldScore scenes — orthogonal to the PermaVid revisit PSNR.