Multi-Image Understanding on MEGABench (test)
54.2AccuracyGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4o2024.12 | 54.2 | |
| Qwen2-VL-7B-Ins.Backbone=7B-Instruct2024.12 | 36 | |
| LLaVA-OV-72BBackbone=72B2024.12 | 33.8 | |
| MAmmoTH-VL-8BBackbone=8B2024.12 | 28.2 | |
| InternVL-2-8BBackbone=8B2024.12 | 27.7 | |
| LLaVA-OV-7BBackbone=7B2024.12 | 23.9 | |
| LLaVA-OV-7Bvariant=Single-Image (SI), Backbone=7B2024.12 | 22.1 |