Audio-Visual Understanding on AV-SpeakerBench
75.1ScoreGemini-3.1 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-3.1 Pro2026.04 | 75.1 | |
| Qwen3.5-Omni-Plus2026.04 | 71.3 | |
| Qwen3.5-Omni-Flash2026.04 | 65.2 | |
| CogniRoute2026.06 | 56.2 | |
| D-ORCAModel Parameters=8B2026.02 | 55 | |
| Qwen3-Omni-InstructModel Parameters=30B-A3B2026.02 | 54.1 | |
| Qwen3 Omni 30BParameters=30B2026.06 | 54.1 | |
| video-SALMONN 2+Model Parameters=7B2026.02 | 51.6 | |
| AVoCaDOModel Parameters=7B2026.02 | 51.6 | |
| Qwen2.5-OmniModel Parameters=7B2026.02 | 42.3 | |
| ARC-Qwen-Video-NarratorModel Parameters=7B2026.02 | 40.2 |