Text-to-Video Generation on VBench & UniBench Dataset
97.44Background ConsistencyUnityVideo
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| UnityVideoConfiguration=Unified ControGen, T2V, and Estimation, Mode=T2V joint generation2025.12 | 97.44 | 64.12 | 23.57 | 47.76 | |
| Wan2.1-14BTask=Text2Video2025.12 | 96.78 | 63.66 | 21.53 | 34.31 | |
| OpenSora2Task=Text2Video2025.12 | 96.51 | 61.51 | 19.87 | 34.48 | |
| HunyuanVideo-13BTask=Text2Video2025.12 | 96.28 | 53.45 | 22.61 | 41.18 | |
| Kling1.6Task=Text2Video2025.12 | 95.33 | 60.48 | 21.76 | 47.05 | |
| AetherTask=Text2Video2025.12 | 95.28 | 48.25 | 20.26 | 37.32 |