Audio-Visual Understanding on Video-Holmes (Score)
0.541ScoreQwen3-Omni-Instruct
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-Omni-InstructModel Parameters=30B-A3B2026.02 | 0.541 | |
| D-ORCAModel Parameters=8B2026.02 | 0.485 | |
| AVoCaDOModel Parameters=7B2026.02 | 0.472 | |
| video-SALMONN 2+Model Parameters=7B2026.02 | 0.469 | |
| Qwen2.5-OmniModel Parameters=7B2026.02 | 0.437 | |
| ARC-Qwen-Video-NarratorModel Parameters=7B2026.02 | 0.432 |