VEFX-Bench - Edit Exclusivity: leaderboard
Metric: Edit-exclusivity score (unrequested changes avoided) as a coverage-adjusted model-level estimate (inverse-propensity-weighted mixed-effects) on the 1-4 rubric scale over the 300 curated VEFX-Bench source-video and instruction pairs, each system's edited video scored by the authors' VEFX-Reward-32B evaluator from soft expected predictions; higher is better. Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Grok Imagine | 3.38 |
| 2 | UniVideo | 3.09 |
| 3 | Seedance 2.0 | 3.09 |
| 4 | Kling 3.0 Omni | 3.04 |
| 5 | Kling o1 | 2.98 |
| 6 | Runway Gen-4.5 | 2.92 |
| 7 | Luma ray 3 | 2.71 |
| 8 | Wan 2.6 | 2.45 |
| 9 | Luma ray 2 | 1.36 |
| 10 | VACE (VEFX-Bench checkpoint unspecified) | 1.18 |
Interactive version: theaggregate.ai/benchmark?slug=vefx-bench-edit-exclusivity · How It Works · Data refreshed daily, snapshot 2026-10-07.