| Benchmark | Model | Rank | Results |
|---|---|---|---|
| image-captioning-on-nocaps-val | MetaLM | #2 | CIDEr: 58.7SPICE: 8.6 |
| image-captioning-on-flickr30k-captions-test | MetaLM | #3 | CIDEr: 43.3SPICE: 11.7 |
| visual-question-answering-on-vqa-v2-val | MetaLM | #9 | Accuracy: 41.1 |
| visual-question-answering-on-ok-vqa | MetaLM | #36 | Accuracy: 11.4 |