VBench
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Text-to-Video Generation | VBench | Quality Score86.73 | 111 | |
| Video generation | VBench | Quality Score86.67 | 102 | |
| Video Generation | VBench 2.0 (test) | Total Score83.79 | 44 | |
| Video Generation | VBench 5s | Total Score84.87 | 35 | |
| Video Generation | VBench (test) | Semantic Score83.4 | 35 | |
| text-to-video generation | VBench HunyuanVideo (test) | VBench Score (%)81.4 | 21 | |
| Video Generation | VBench 1.0 (test) | Image Quality84.71 | 21 | |
| Video Generation | VBench short video (test) | Subject Consistency97.79 | 16 | |
| Video Generation | VBench aesthetic and imaging quality dimensions | Aesthetic Quality0.6499 | 15 | |
| Text-to-Video Generation | VBench 2024 (test) | Total Score81.01 | 15 | |
| Text-to-Video Generation | VBench (test) | Total Score81.4 | 14 | |
| Video Generation | VBench Long | Overall Quality Score84.11 | 14 | |
| Video Generation | VBench v1 (test) | Latency (s)7.39 | 13 | |
| Text-to-Video Generation | VBench 1.0 (test) | VBench Score82.1 | 13 | |
| Image-to-Video Generation | VBench I2V 1.0 (test) | Subject Consistency98.76 | 13 | |
| Video Generation | VBench Leaderboard Comparison 1.0 | Total Score84.7 | 12 | |
| Instance Insertion | VBench official (test) | Background Consistency94.63 | 12 | |
| Text-to-Video Generation | VBench | Evaluation Time (min)6 | 12 | |
| Short Video Generation | VBench official prompts | Total Score84.26 | 11 | |
| Text-to-Video generation | VBench 17 frames, 512x512 | UR45.9 | 11 | |
| Video Generation | VBench Overall | Throughput (FPS)17 | 11 | |
| Short Video Generation | VBench 2024 | Total Score85.12 | 11 | |
| Video Generation | VBench 30-second generation | Imaging Quality85.52 | 11 | |
| Video Generation | VBench Custom | Subject Consistency97.61 | 11 | |
| Text-to-Video | VBench 2.0 (test) | Composition70.8 | 10 |
Showing 25 of 111 rows