| multi-task-language-understanding-on-livebench-2024-11-25 | Qwen3-VL-235B-A22B-Thinking | #1 | Score: 79.6 |
| text-generation-on-writingbench | Qwen3-VL-235B-A22B-Thinking | #1 | Score (0-10): 8.67 |
| visual-question-answering-on-docvqa-test | Qwen3-VL-235B-A22B-Instruct | #1 | ANLS: 0.971 |
| visual-question-answering-vqa-on-infographicvqa | Qwen3-VL-235B-A22B-Thinking | #1 | ANLS: 89.5 |
| chart-question-answering-on-charxiv-dq | Qwen3-VL-235B-A22B-Thinking | #2 | Accuracy: 90.5 |
| instruction-following-on-arena-hard-auto-v2 | Qwen3-VL-235B-A22B-Instruct | #2 | Accuracy: 77.4 |
| optical-character-recognition-ocr-on-ocrbench-v2 | Qwen3-VL-32B-Thinking | #2 | English Score: 68.4 |
| optical-character-recognition-ocr-on-ocrbench-v2 | Qwen3-VL-235B-A22B-Thinking | #8 | Chinese Score: 63.5 |
| visual-question-answering-vqa-on-ai2d-test-with-mask | Qwen3-VL-235B-A22B-Instruct | #2 | Accuracy: 89.7 |
| instruction-following-on-multi-if | Qwen3-VL (235B-A22B, Thinking) | #3 | Accuracy: 79.1 |
| optical-character-recognition-ocr-on-ocrbench | Qwen3-VL-235B-A22B-Instruct | #3 | Score: 920 |
| optical-character-recognition-ocr-on-omnidocbench-v1-0 | Qwen3-VL-235B-A22B-Instruct | #3 | Edit Distance: 0.143 |
| visual-grounding-on-refcoco-refcoco-refcocog-overall | Qwen3-VL-235B-A22B-Thinking | #3 | Accuracy: 92.1 |
| visual-question-answering-vqa-on-mmbench-cn | Qwen3-VL-235B-A22B-Instruct | #3 | Accuracy: 88.9 |
| computer-use-agents-on-osw-g-osworld-g | Qwen3-VL-235B-A22B-Thinking | #4 | Accuracy: 68.3 |
| multi-task-language-understanding-on-mmlu-redux | Qwen3-VL-235B-A22B-Thinking | #4 | Accuracy: 93.7 |
| visual-question-answering-vqa-on-simplevqa | Qwen3-VL-235B-A22B-Instruct | #4 | Accuracy: 63.0 |
| zero-shot-object-detection-on-odinw13 | Qwen3-VL-235B-A22B-Instruct (zero-shot) | #4 | mAP: 48.6 |
| multi-task-language-understanding-on-include | Qwen3-VL-235B-A22B-Thinking | #5 | Accuracy: 80.0 |
| multimodal-reasoning-on-zerobench | Qwen3-VL-235B-A22B-Thinking | #6 | Accuracy: 4 |
| video-question-answering-on-mvbench | Qwen3-VL-235B-A22B-Instruct | #6 | Avg.: 76.5 |
| visual-question-answering-vqa-on-blink | Qwen3-VL-235B-A22B-Instruct | #6 | Accuracy: 70.7 |
| zero-shot-video-question-answer-on-video-mme-w-o-subs | Qwen3-VL-235B-A22B-Instruct | #6 | Accuracy (%): 79.2 |
| chart-question-answering-on-chartqa | Qwen3-VL-235B-A22B-Thinking | #7 | 1:1 Accuracy: 90.3 |
| multiple-choice-question-answering-mcqa-on-supergpqa | Qwen3-VL-235B-A22B-Thinking | #7 | Accuracy: 64.3 |
| visual-question-answering-vqa-on-hallusionbench-average-accuracy | Qwen3-VL-235B-A22B-Thinking | #7 | Accuracy: 66.7 |
| multimodal-reasoning-on-mmmu | Qwen3-VL-235B-A22B-Thinking | #9 | Accuracy: 80.6 |
| video-question-answering-on-mlvu | Qwen3-VL-235B-A22B-Instruct | #9 | Accuracy: 84.3 |
| multimodal-reasoning-on-mathvista | Qwen3-VL-235B-A22B-Thinking | #10 | Accuracy: 85.8 |
| object-counting-on-countbench | Qwen3-VL-235B-A22B-Thinking | #11 | Accuracy: 93.7 |
| video-question-answering-on-mmvu | Qwen3-VL-235B-A22B-Thinking | #11 | Accuracy: 71.1 |
| video-question-answering-on-video-mmmu | Qwen3-VL-235B-A22B-Thinking | #11 | Accuracy: 80.0 |
| computer-use-agents-on-screenspot-pro | Qwen3-VL-235B-A22B-Instruct | #12 | Accuracy: 62.0 |
| vision-language-understanding-on-mmbench-v1-1 | Qwen3-VL-8B-Instruct | #12 | Accuracy: 85.0 |
| visual-question-answering-vqa-on-realworldqa | Qwen3-VL-235B-A22B-Thinking | #12 | Accuracy: 81.3 |
| vision-language-understanding-on-mmstar | Qwen3-VL-235B-A22B-Thinking | #13 | Accuracy: 78.7 |
| chart-question-answering-on-charxiv-rq | Qwen3-VL-235B-A22B-Thinking | #14 | Accuracy: 66.1 |
| multimodal-reasoning-on-mathvision | Qwen3-VL-235B-A22B-Thinking | #15 | Accuracy: 74.6 |
| video-question-answering-on-lvbench | Qwen3-VL-235B-A22B-Instruct | #15 | Accuracy: 67.7 |
| visual-question-answering-vqa-on-mmbench-en | Qwen3-VL-4B-Instruct | #18 | Accuracy: 83.9 |
| computer-use-agents-on-androidworld | Qwen3-VL-235B-A22B-Instruct | #19 | Success Rate: 63.7 |
| multimodal-reasoning-on-mmmu-pro | Qwen3-VL-235B-A22B-Thinking | #20 | Accuracy: 69.3 |
| mathematical-reasoning-on-aime-2025 | Qwen3-VL-235B-A22B-Thinking | #24 | Accuracy: 89.7 |
| multi-task-language-understanding-on-mmlu-pro | Qwen3-VL-235B-A22B-Thinking | #25 | Accuracy: 83.8 |
| code-generation-on-livecodebench-v6 | Qwen3-VL-235B-A22B-Thinking | #33 | Pass@1: 70.1 |