Visual Question Answering on VQAv2 (dev)
84.19AccuracyBEiT-3
Evaluation Results
| Method | Links | |
|---|---|---|
| BEiT-3Example=21M2023.06 | 84.19 | |
| CoCaExample=4.8B2023.06 | 82.3 | |
| BLIP-2Example=129M2023.06 | 82.19 | |
| OFAExample=18M2023.06 | 82 | |
| FlamingoExample=2.3B2023.06 | 82 | |
| GIT2Example=12.9B2023.06 | 81.7 | |
| mPLUG-2Example=417M2023.06 | 81.11 | |
| COSAExample=415M, parameters=1.2B2023.06 | 80.46 | |
| FlorenceExample=900M2023.06 | 80.16 | |
| SimVLMExample=1.8B2023.06 | 80.03 | |
| GITExample=0.8B2023.06 | 78.6 | |
| VALORExample=433.5M2023.06 | 78.46 | |
| BLIPExample=129M2023.06 | 78.25 | |
| Unified-IO2023.06 | 77.9 | |
| ShikraModel Size=13B2023.06 | 77.36 | |
| BLIP-22023.06 | 65 | |
| FM-80BModel Size=80B2023.06 | 56.3 | |
| FM-9BModel Size=9B2023.06 | 51.8 | |
| Kosmos-12023.06 | 51 |