| temporal-relation-extraction-on-vinoground | Qwen2-VL-72B | #1 | Text Score: 50.4Video Score: 32.6Group Score: 17.4 |
| temporal-relation-extraction-on-vinoground | Qwen2-VL-7B | #4 | Text Score: 40.2Video Score: 32.4Group Score: 15.2 |
| video-question-answering-on-implicitqa | Qwen2 VL - 7B | #2 | Average Accuracy: 44.9Macro Average Accuracy: 46.0 |
| zero-shot-video-question-answer-on-egoschema-fullset | Qwen2-VL-72B | #2 | Accuracy: 77.9 |
| zero-shot-video-question-answer-on-egoschema-fullset | Qwen2-VL-7B | #14 | Accuracy: 66.7 |
| zero-shot-video-question-answer-on-egoschema-fullset | Qwen2-VL-2B | #38 | Accuracy: 54.9 |
| visual-question-answering-on-docvqa-test | Qwen2-VL-72B | #3 | ANLS: 0.965 |
| visual-question-answering-on-docvqa-test | Qwen2-VL-7B | #15 | ANLS: 0.945 |
| visual-question-answering-on-docvqa-test | Qwen2-VL-2B | #43 | ANLS: 0.901 |
| visual-question-answering-vqa-on-mme | Qwen2-VL-72B | #3 | Score: 2482.7 |
| visual-question-answering-vqa-on-mme | Qwen2-VL-7B | #10 | Score: 2326.8 |
| visual-question-answering-vqa-on-mme | Qwen2-VL-2B | #20 | Score: 1872.0 |
| visual-question-answering-vqa-on-vlm2-bench | Qwen2-VL-7B | #3 | GC-mat: 27.80GC-trk: 19.18OC-cpr: 68.06OC-cnt: 45.99… |
| visual-question-answering-on-mm-vet-v2 | Qwen2-VL-72B (qwen-vl-max-0809) | #4 | GPT-4 score: 66.9±0.3Params: 72B |
| visual-question-answering-vqa-on-mtvqa | Qwen2-VL-72B | #4 | Accuracy: 30.9 |
| visual-question-answering-vqa-on-mtvqa | Qwen2-VL-7B | #5 | Accuracy: 25.6 |
| visual-question-answering-vqa-on-mtvqa | Qwen2-VL-2B | #6 | Accuracy: 18.1 |
| video-question-answering-on-ovbench | Qwen2-VL (7B) | #5 | AVG: 49.7 |
| zero-shot-video-question-answer-on-vnbench | Qwen2-VL-7B | #5 | Accuracy: 33.9 |
| video-question-answering-on-tvbench | Qwen2-VL-72B | #6 | Average Accuracy: 52.7 |
| video-question-answering-on-tvbench | Qwen2-VL-7B | #13 | Average Accuracy: 43.8 |
| visual-question-answering-on-mm-vet | Qwen2-VL-72B | #6 | GPT-4 score: 74.0 |
| visual-question-answering-on-mm-vet | Qwen2-VL-7B | #38 | GPT-4 score: 62.0 |
| visual-question-answering-on-mm-vet | Qwen2-VL-2B | #70 | GPT-4 score: 49.5 |
| visual-grounding-on-refcoco-refcoco-refcocog-overall | Qwen2-VL-72B | #8 | Accuracy: 91.07 |
| visual-question-answering-vqa-on-mmbench-cn | Qwen2-VL-72B | #8 | Accuracy: 86.6 |
| visual-question-answering-vqa-on-mmbench-cn | Qwen2-VL-7B | #13 | Accuracy: 80.5 |
| visual-question-answering-vqa-on-mmbench-cn | Qwen2-VL-2B | #18 | Accuracy: 73.5 |
| video-question-answering-on-next-qa | Qwen2-VL(7B) | #9 | Accuracy: 81.2 |
| vision-language-understanding-on-mmbench-v1-1 | Qwen2-VL-72B | #11 | Accuracy: 85.9 |
| vision-language-understanding-on-mmbench-v1-1 | Qwen2-VL-7B | #20 | Accuracy: 80.7 |
| vision-language-understanding-on-mmbench-v1-1 | Qwen2-VL-2B | #24 | Accuracy: 72.2 |
| visual-question-answering-vqa-on-mmbench-en | Qwen2-VL-72B | #11 | Accuracy: 86.5 |
| visual-question-answering-vqa-on-mmbench-en | Qwen2-VL-7B | #21 | Accuracy: 83.0 |
| visual-question-answering-vqa-on-mmbench-en | Qwen2-VL-2B | #30 | Accuracy: 74.9 |
| video-question-answering-on-video-mme-with-subtitles | Qwen2-VL-72B (w/ subtitles) | #15 | Accuracy: 77.8 |
| optical-character-recognition-ocr-on-ocrbench | Qwen2-VL-72B | #16 | Score: 877 |
| visual-question-answering-vqa-on-hallusionbench-average-accuracy | Qwen2-VL-72B | #17 | Accuracy: 58.1 |
| visual-question-answering-vqa-on-hallusionbench-average-accuracy | Qwen2-VL-7B | #19 | Accuracy: 50.6 |
| visual-question-answering-vqa-on-hallusionbench-average-accuracy | Qwen2-VL-2B | #22 | Accuracy: 41.7 |
| visual-question-answering-vqa-on-realworldqa | Qwen2-VL-72B | #17 | Accuracy: 77.8 |
| visual-question-answering-vqa-on-realworldqa | Qwen2-VL-7B | #53 | Accuracy: 70.1 |
| visual-question-answering-vqa-on-realworldqa | Qwen2-VL-2B | #81 | Accuracy: 62.9 |
| chart-question-answering-on-chartqa | Qwen2-VL-72B | #18 | 1:1 Accuracy: 88.3 |
| chart-question-answering-on-chartqa | Qwen2-VL-7B | #56 | 1:1 Accuracy: 83.0 |
| chart-question-answering-on-chartqa | Qwen2-VL-2B | #89 | 1:1 Accuracy: 73.5 |
| video-question-answering-on-mvbench | Qwen2-VL-72B | #18 | Avg.: 73.6 |
| video-question-answering-on-mvbench | Qwen2-VL-7B | #45 | Avg.: 67.0 |
| video-question-answering-on-mvbench | Qwen2-VL-2B | #59 | Avg.: 63.2 |
| zero-shot-video-question-answer-on-video-mme-w-o-subs | Qwen2-VL-72B | #26 | Accuracy (%): 71.2 |
| zero-shot-video-question-answer-on-video-mme-w-o-subs | Qwen2-VL-7B | #42 | Accuracy (%): 63.3 |
| zero-shot-video-question-answer-on-video-mme-w-o-subs | Qwen2-VL-2B | #53 | Accuracy (%): 55.6 |
| natural-language-visual-grounding-on-screenspot | Qwen2-VL-7B | #31 | Accuracy (%): 42.1 |
| multimodal-reasoning-on-mmmu-pro | Qwen2-VL-72B | #45 | Accuracy: 46.2 |
| multimodal-reasoning-on-mmmu-pro | Qwen2-VL-7B | #47 | Accuracy: 43.5 |
| multimodal-reasoning-on-mmmu-pro | Qwen2-VL-2B | #56 | Accuracy: 37.6 |
| vision-language-understanding-on-mmstar | Qwen2-VL-72B | #51 | Accuracy: 68.3 |
| vision-language-understanding-on-mmstar | Qwen2-VL-7B | #84 | Accuracy: 60.7 |
| vision-language-understanding-on-mmstar | Qwen2-VL-2B | #98 | Accuracy: 48.0 |
| multimodal-reasoning-on-mmmu | Qwen2-VL-72B | #52 | Accuracy: 64.5 |
| multimodal-reasoning-on-mmmu | Qwen2-VL-7B | #83 | Accuracy: 54.1 |
| multimodal-reasoning-on-mmmu | Qwen2-VL-2B | #124 | Accuracy: 41.1 |
| multimodal-reasoning-on-mathvision | Qwen2-VL-72B | #56 | Accuracy: 25.9 |
| multimodal-reasoning-on-mathvision | Qwen2-VL-7B | #62 | Accuracy: 16.3 |
| multimodal-reasoning-on-mathvision | Qwen2-VL-2B | #64 | Accuracy: 12.4 |
| multimodal-reasoning-on-mathvista | Qwen2-VL-72B | #65 | Accuracy: 70.5 |
| multimodal-reasoning-on-mathvista | Qwen2-VL-7B | #101 | Accuracy: 58.2 |
| multimodal-reasoning-on-mathvista | Qwen2-VL-2B | #113 | Accuracy: 43.0 |