VBench-2.0
Submission-based video evaluation; Total and Creativity scores
2.0generative mediaactive
Score in %, VBench-2.0
Definition
- Version
- 2.0
- Category
- generative media
- Maintainer
- Vchitect
- Status
- active
- Home page
- https://huggingface.co/spaces/Vchitect/VBench_Leaderboard
Sources
- VBench Leaderboard - a Hugging Face Space by Vchitect huggingface.co https://huggingface.co/spaces/Vchitect/VBench_Leaderboard
Scores at a Glance
Score in %, VBench-2.0
Higher is better for this unit.
1
Veo 3Google DeepMindTotal score
Vidu Q1ShengShu TechnologyTotal score
SoraOpenAICreativity score; Sora-480p
Seedance 1.0ByteDanceTotal score; Seedance 1.0 Pro
Kling 1.6KuaishouTotal score
Veo 3Google DeepMindTotal score66.72 %
maintainer-run2JT-CVTotal score64.6 %
maintainer-run3ABot-World v0.1Total score64.47 %
maintainer-run4Vidu Q1ShengShu TechnologyTotal score62.7 %
maintainer-run5ToMoviee 2.0Total score61.78 %
maintainer-run6SoraOpenAICreativity score; Sora-480p60.57 %
maintainer-run7Seedance 1.0ByteDanceTotal score; Seedance 1.0 Pro59.81 %
maintainer-run8Kling 1.6KuaishouTotal score59 %
maintainer-run0255075100
Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.
Scores
| Model | Score | Who ran it | Setting | View or population | Date | Source |
|---|---|---|---|---|---|---|
| Veo 3 | 66.72 % | maintainer-run | Total score | rank 1, VBench-2.0 | 4 Sep 2025 | huggingface.co |
| JT-CV | 64.6 % | maintainer-run | Total score | rank 2, VBench-2.0 | 1 Apr 2026 | huggingface.co |
| ABot-World v0.1 | 64.47 % | maintainer-run | Total score | rank 3, VBench-2.0 | 21 Mar 2026 | huggingface.co |
| Vidu Q1 | 62.7 % | maintainer-run | Total score | rank 4, VBench-2.0 | 21 Apr 2025 | huggingface.co |
| ToMoviee 2.0 | 61.78 % | maintainer-run | Total score | rank 5, VBench-2.0 | 18 Sep 2025 | huggingface.co |
| Veo 3 | 60.85 % | maintainer-run | Creativity score | VBench-2.0 | 4 Sep 2025 | huggingface.co |
| Sora | 60.57 % | maintainer-run | Creativity score; Sora-480p | VBench-2.0 | 28 Mar 2025 | huggingface.co |
| Seedance 1.0 | 59.81 % | maintainer-run | Total score; Seedance 1.0 Pro | rank 7, VBench-2.0 | 26 Jun 2025 | huggingface.co |
| Kling 1.6 | 59 % | maintainer-run | Total score | rank 8, VBench-2.0 | 28 Mar 2025 | huggingface.co |
| Sora | 58.38 % | maintainer-run | Total score; Sora-480p | rank 9, VBench-2.0 | 28 Mar 2025 | huggingface.co |
| Vidu Q1 | 56.54 % | maintainer-run | Creativity score | VBench-2.0 | 21 Apr 2025 | huggingface.co |
| ABot-World v0.1 | 55.29 % | maintainer-run | Creativity score | VBench-2.0 | 21 Mar 2026 | huggingface.co |
| Seedance 1.0 | 53.04 % | maintainer-run | Creativity score; Seedance 1.0 Pro | VBench-2.0 | 26 Jun 2025 | huggingface.co |
| JT-CV | 51.8 % | maintainer-run | Creativity score | VBench-2.0 | 1 Apr 2026 | huggingface.co |
| Kling 1.6 | 48.58 % | maintainer-run | Creativity score | VBench-2.0 | 28 Mar 2025 | huggingface.co |
| ToMoviee 2.0 | 45.96 % | maintainer-run | Creativity score | VBench-2.0 | 18 Sep 2025 | huggingface.co |
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "VBench-2.0", https://aurion.semai.id/benchmarks/vbench-2-0/, records of October 2026.