Audio-Visual Video Understanding on OmniVideoBench truncated <= 5min
57.6ScoreOurs(Qwen3-Omni)
Evaluation Results
| Method | Links | |
|---|---|---|
| Ours(Qwen3-Omni)Size=30B2026.06 | 57.6 | |
| Ours(VITA-1.5)Size=8B2026.06 | 56.35 | |
| Ours(Qwen2.5-Omni)Size=7B2026.06 | 55 | |
| Qwen3-OmniSize=30B2026.06 | 53.44 | |
| OmniVinciSize=9B2026.06 | 52.81 | |
| Uni-MoE-2.0-OmniSize=30B2026.06 | 52.81 | |
| MiniCPM-o 4.5Size=9B2026.06 | 52.29 | |
| video-SALMONN 2+Size=7B2026.06 | 51.88 | |
| Qwen2.5-OmniSize=7B2026.06 | 48.85 | |
| VITA-1.5Size=8B2026.06 | 48.65 | |
| Ours(Qwen3-Omni)Size=30B2026.06 | 44.81 | |
| Qwen3-OmniSize=30B2026.06 | 43.84 | |
| MiniCPM-o 4.5Size=9B2026.06 | 41.06 | |
| Ours(Qwen2.5-Omni)Size=7B2026.06 | 39.88 | |
| Uni-MoE-2.0-OmniSize=30B2026.06 | 38.55 | |
| Ours(VITA-1.5)Size=8B2026.06 | 36.94 | |
| OmniVinciSize=9B2026.06 | 36.74 | |
| Qwen2.5-OmniSize=7B2026.06 | 36.54 | |
| VITA-1.5Size=8B2026.06 | 36.35 | |
| video-SALMONN 2+Size=7B2026.06 | 35.36 |