| Benchmark | Model | Rank | Results |
|---|---|---|---|
| visual-question-answering-vqa-on-textvqa-val | Idefics3-8B | #20 | Accuracy: 74.9 |
| visual-question-answering-on-docvqa-test | Idefics3-8B | #53 | ANLS: 0.877 |
| vision-language-understanding-on-mmstar | Idefics3-8B | #88 | Accuracy: 55.9 |
| multimodal-reasoning-on-mathvista | Idefics3-8B | #97 | Accuracy: 58.4 |
| multimodal-reasoning-on-mmmu | Idefics3-8B | #105 | Accuracy: 46.6 |