Paralinguistic Perception on VoxParadox
35Age ScoreAudio Flamingo 2 (AF2)
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Audio Flamingo 2 (AF2)Model Type=Open-Source Audio LLMs2026.05 | 35 | 99 | 0 | 19.5 | 19 | 21 | 18 | 38.5 | 27.5 | 31 | 30.85 | |
| Qwen3-OmniModel Type=Open-Source Omni LLMs2026.05 | 17.5 | 4.5 | 0.5 | 4 | 7 | 4 | 2.5 | 0 | 12 | 54 | 10.6 | |
| Gemini 2.5 FlashModel Type=Closed-Source Models2026.05 | 15 | 14.5 | 6 | 19 | 31 | 16 | 13 | 19 | 21 | 92.5 | 24.7 | |
| Step-Audio-R1Model Type=Open-Source Audio LLMs2026.05 | 10.5 | 12.5 | 1.5 | 11.5 | 4.5 | 9.5 | 9.5 | 13 | 9 | 93 | 17.45 | |
| Audio Flamingo 3 (AF3)Model Type=Open-Source Audio LLMs2026.05 | 10 | 16 | 24.5 | 11 | 11.5 | 11.5 | 9.5 | 34 | 23.5 | 22.5 | 17.4 | |
| SALMONN-7BModel Type=Open-Source Audio LLMs2026.05 | 10 | 12.5 | 0 | 0 | 0 | 17 | 0 | 21.5 | 0 | 0 | 6.1 | |
| Kimi-Audio-7B-InstructModel Type=Open-Source Audio LLMs2026.05 | 9 | 24.5 | 79 | 7.5 | 12.5 | 11 | 8 | 5 | 20.5 | 13 | 19 | |
| MiMo-Audio-7B-InstructModel Type=Open-Source Audio LLMs2026.05 | 7.5 | 50 | 0.5 | 11.5 | 19 | 15 | 15 | 11.5 | 30 | 36 | 19.6 | |
| GPT-4o AudioModel Type=Closed-Source Models2026.05 | 4.5 | 1 | 0 | 7 | 5 | 8 | 5 | 0.5 | 27 | 28 | 8.6 | |
| VITA-AudioModel Type=Open-Source Audio LLMs2026.05 | 3.5 | 3.5 | 0 | 8.5 | 8 | 16 | 9 | 0 | 12 | 8 | 6.85 | |
| Qwen2-Audio-7B-InstructModel Type=Open-Source Audio LLMs2026.05 | 2 | 3 | 14 | 26.5 | 22 | 24.5 | 19 | 0.5 | 24.5 | 12.5 | 14.85 | |
| Qwen2.5-Omni-7BModel Type=Open-Source Omni LLMs2026.05 | 1.5 | 2 | 3 | 13 | 8 | 8 | 12.5 | 4.5 | 15.5 | 11.5 | 7.95 |