Audio-Visual Question Answering on Video-MME
75ScoreGemini 2.5 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini 2.5 ProModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Approach=caption-to-QA cascade2025.10 | 75 | |
| Gemini 2.5 FlashModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Approach=caption-to-QA cascade2025.10 | 69.1 | |
| Omni-Captioner-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Approach=caption-to-QA cascade2025.10 | 67.1 | |
| video-SALMONN 2-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Approach=caption-to-QA cascade2025.10 | 65.9 | |
| Gemini 2.0 FlashModel Category=Proprietary Models, QA Backbone=GPT-4o, Evaluation Approach=caption-to-QA cascade2025.10 | 64.4 | |
| Qwen2.5-Omni-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Approach=caption-to-QA cascade2025.10 | 52.7 | |
| VideoLLaMA 2-7BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Approach=caption-to-QA cascade2025.10 | 44.4 | |
| video-SALMONN-13BModel Category=Open-Source Models, QA Backbone=GPT-4o, Evaluation Approach=caption-to-QA cascade2025.10 | 41.8 |