Video-to-Video Generation on Simple-V2V Bench V6
78.98Visual Text Alignment ScoreSeedream 5.0 Lite
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Seedream 5.0 LiteScoring Protocol=dual-dimension scoring (min(Q, A) × 10)2026.05 | 78.98 | 68.67 | 74.66 | 88.79 | 48.71 | 16.78 | 32.42 | 58.43 | |
| GPT Image 2Scoring Protocol=dual-dimension scoring (min(Q, A) × 10)2026.05 | 78.3 | 92.39 | 75.8 | 91.82 | 60.34 | 20 | 33.98 | 64.66 | |
| Nano Banana 2Scoring Protocol=dual-dimension scoring (min(Q, A) × 10)2026.05 | 59.2 | 69.66 | 77.95 | 67.05 | 44.66 | 19.09 | 22.27 | 51.41 | |
| BAGEL-7B-MoTScoring Protocol=dual-dimension scoring (min(Q, A) × 10)2026.05 | 43.52 | 10 | 11.93 | 10.34 | 10.23 | 10 | 10.57 | 15.23 | |
| V2V-ZeroScoring Protocol=dual-dimension scoring (min(Q, A) × 10)2026.05 | 34.77 | 76.93 | 42.84 | 23.98 | 20.34 | 13.3 | 16.59 | 32.68 | |
| HunyuanVideo-1.5Scoring Protocol=dual-dimension scoring (min(Q, A) × 10), Evaluation Method=VLM judge applied to generated video directly2026.05 | 17.73 | 32.5 | 25.68 | 19.2 | 17.27 | 12.39 | 16.25 | 20.15 | |
| Qwen-Image-Edit-2511Scoring Protocol=dual-dimension scoring (min(Q, A) × 10)2026.05 | 15.68 | 16.93 | 34.2 | 23.18 | 17.05 | 13.41 | 17.16 | 19.66 |