| zero-shot-video-question-answer-on-egoschema-fullset | Kimi-VL-A3B-Instruct | #1 | Accuracy: 78.5 |
| visual-question-answering-on-mm-vet | Kimi-VL-A3B-Thinking-2506 | #3 | GPT-4 score: 78.1 |
| visual-question-answering-on-mm-vet | Kimi-VL-A3B-Instruct | #18 | GPT-4 score: 66.7 |
| visual-question-answering-vqa-on-infographicvqa | Kimi-VL-A3B-Instruct | #5 | ANLS: 83.2 |
| video-question-answering-on-tomato | Kimi-VL-A3B-Instruct | #11 | Accuracy: 31.7 |
| vision-language-understanding-on-mmbench-v1-1 | Kimi-VL-A3B-Thinking-2506 | #14 | Accuracy: 84.4 |
| vision-language-understanding-on-mmbench-v1-1 | Kimi-VL-A3B-Instruct | #16 | Accuracy: 83.1 |
| video-question-answering-on-video-mme-with-subtitles | Kimi-VL-A3B-Instruct (w/ subs) | #20 | Accuracy: 72.6 |
| video-question-answering-on-video-mme-with-subtitles | Kimi-VL-A3B-Thinking-2506 (w/ subs) | #23 | Accuracy: 71.9 |
| optical-character-recognition-ocr-on-ocrbench | Kimi-VL-A3B-Thinking-2506 | #21 | Score: 869 |
| video-question-answering-on-video-mmmu | Kimi-VL-A3B-Thinking-2506 | #21 | Accuracy: 65.2 |
| video-question-answering-on-video-mmmu | Kimi-VL-A3B-Instruct | #28 | Accuracy: 52.6 |
| computer-use-agents-on-screenspot-v2 | Kimi-VL-A3B-Instruct | #23 | Accuracy: 92.8 |
| multimodal-reasoning-on-mathvision | Kimi-VL-A3B-Thinking-2506 | #25 | Accuracy: 56.9 |
| multimodal-reasoning-on-mathvision | Kimi-VL-A3B-Instruct | #59 | Accuracy: 21.4 |
| video-question-answering-on-mlvu | Kimi-VL-A3B-Instruct | #25 | Accuracy: 74.2 |
| video-question-answering-on-mmvu | Kimi-VL-A3B-Thinking-2506 | #26 | Accuracy: 57.5 |
| video-question-answering-on-mmvu | Kimi-VL-A3B-Instruct | #30 | Accuracy: 52.2 |
| computer-use-agents-on-screenspot-pro | Kimi-VL-A3B-Thinking-2506 | #29 | Accuracy: 52.8 |
| multimodal-reasoning-on-mathvista | Kimi-VL-A3B-Thinking-2506 | #29 | Accuracy: 80.1 |
| multimodal-reasoning-on-mathvista | Kimi-VL-A3B-Instruct | #67 | Accuracy: 68.7 |
| zero-shot-video-question-answer-on-video-mme-w-o-subs | Kimi-VL-A3B-Instruct | #29 | Accuracy (%): 67.8 |
| visual-question-answering-vqa-on-blink | Kimi-VL-A3B-Instruct | #30 | Accuracy: 57.3 |
| vision-language-understanding-on-mmstar | Kimi-VL-A3B-Thinking-2506 | #40 | Accuracy: 70.4 |
| vision-language-understanding-on-mmstar | Kimi-VL-A3B-Instruct | #78 | Accuracy: 61.3 |
| multimodal-reasoning-on-mmmu-pro | Kimi-VL-A3B-Thinking-2506 | #43 | Accuracy: 46.3 |
| multimodal-reasoning-on-mmmu | Kimi-VL-A3B-Thinking-2506 | #51 | Accuracy: 64.0 |
| multimodal-reasoning-on-mmmu | Kimi-VL-A3B-Instruct | #66 | Accuracy: 57.0 |
| visual-question-answering-vqa-on-realworldqa | Kimi-VL-A3B-Thinking-2506 | #52 | Accuracy: 70.0 |
| visual-question-answering-vqa-on-realworldqa | Kimi-VL-A3B-Instruct | #57 | Accuracy: 68.1 |