Multimodal Cloze on Omni-Cloze Audio
53.2AccuracyAudio-Captioner-7B
Evaluation Results
| Method | Links | |
|---|---|---|
| Audio-Captioner-7BModel Type=Open-Source Models2025.10 | 53.2 | |
| Audio-Captioner-7BModel Type=Open-Source Models2025.10 | 53.2 | |
| Gemini 2.5 ProModel Type=Proprietary Models2025.10 | 48 | |
| Gemini 2.5 ProModel Type=Proprietary Models2025.10 | 48 | |
| Gemini 2.5 FlashModel Type=Proprietary Models2025.10 | 42.6 | |
| Gemini 2.5 FlashModel Type=Proprietary Models2025.10 | 42.6 | |
| GPT-4o AudioModel Type=Proprietary Models2025.10 | 35.8 | |
| GPT-4o AudioModel Type=Proprietary Models2025.10 | 35.8 | |
| Qwen2.5-Omni-7BModel Type=Open-Source Models2025.10 | 25.8 | |
| Qwen2.5-Omni-7BModel Type=Open-Source Models2025.10 | 25.8 | |
| Qwen2-Audio-7BModel Type=Open-Source Models2025.10 | 22.2 | |
| Qwen2-Audio-7BModel Type=Open-Source Models2025.10 | 22.2 | |
| Gemini 2.0 FlashModel Type=Proprietary Models2025.10 | 20 | |
| Gemini 2.0 FlashModel Type=Proprietary Models2025.10 | 20 | |
| MiDashengLM-7BModel Type=Open-Source Models2025.10 | 19.5 | |
| MiDashengLM-7BModel Type=Open-Source Models2025.10 | 19.5 | |
| SALMONN-13BModel Type=Open-Source Models2025.10 | 10.6 | |
| SALMONN-13BModel Type=Open-Source Models2025.10 | 10.6 |