Multi-choice Question Answering on M3GIA
59.8AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4oModel Modality Type=Proprietary2024.10 | 59.8 | |
| Qwen2 VLModel Scale=7B, Model Modality Type=Vision-language2024.10 | 37.3 | |
| Baichuan-omniModel Scale=7B, Model Modality Type=Omni-modal2024.10 | 34.7 | |
| MiniCPM-Llama3-V 2.5Model Scale=8B, Model Modality Type=Vision-language2024.10 | 30.3 | |
| VITAModel Scale=8x7B, Model Modality Type=Omni-modal2024.10 | 27.7 |