| science-question-answering-on-scienceqa | Phi-4-Multimodal | #2 | Avg. Accuracy: 97.5 |
| math-word-problem-solving-on-mgsm-0-shot | Phi-4-Mini (0-shot, CoT) | #5 | Accuracy: 63.9 |
| visual-question-answering-vqa-on-infographicvqa | Phi-4-Multimodal | #12 | ANLS: 72.7 |
| multi-task-language-understanding-on-mmmlu | Phi-4-Mini (5-shot) | #15 | Accuracy: 49.3 |
| visual-question-answering-on-docvqa-test | Phi-4-Multimodal | #22 | ANLS: 0.932 |
| object-hallucination-evaluation-on-pope | Phi-4-Multimodal | #27 | F1: 85.6 |
| multi-task-language-understanding-on-mmlu | Phi-4-Mini (5-shot) | #35 | Average (%): 67.3 |
| optical-character-recognition-ocr-on-ocrbench | Phi-4-Multimodal | #40 | Score: 844 |
| question-answering-on-boolq | Phi-4-Mini (2-shot) | #41 | Accuracy: 81.2 |
| math-word-problem-solving-on-math | Phi-4-Mini (0-shot, CoT) | #44 | Accuracy: 64.0 |
| multimodal-reasoning-on-mmmu-pro | Phi-4-Multimodal | #53 | Accuracy: 38.5 |
| mathematical-reasoning-on-math-500 | Phi-4-Mini reasoning | #57 | Accuracy: 90.4 |
| chart-question-answering-on-chartqa | Phi-4-Multimodal | #65 | 1:1 Accuracy: 81.4 |
| code-generation-on-mbpp | Phi-4-Mini (3-shot) | #75 | Accuracy: 65.3 |
| multimodal-reasoning-on-mmmu | Phi-4-Multimodal | #75 | Accuracy: 55.1 |
| code-generation-on-humaneval | Phi-4-Mini | #76 | Pass@1: 74.4 |
| question-answering-on-piqa | Phi-4-Mini (5-shot) | #82 | Accuracy: 77.6 |
| mathematical-reasoning-on-aime24 | Phi-4-Mini reasoning | #84 | Acc: 50.0 |
| arithmetic-reasoning-on-gsm8k | Phi-4-Mini (8-shot, CoT) | #89 | Accuracy: 88.6 |
| common-sense-reasoning-on-winogrande | Phi-4-Mini (5-shot) | #89 | Accuracy: 67.0 |
| multimodal-reasoning-on-mathvista | Phi-4-Multimodal | #91 | Accuracy: 62.4 |
| multi-task-language-understanding-on-mmlu-pro | Phi-4-Mini (0-shot, CoT) | #112 | Accuracy: 52.8 |
| sentence-completion-on-hellaswag | Phi-4-Mini (5-shot) | #113 | Accuracy: 69.1 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | Phi-4-Mini reasoning | #118 | Accuracy: 49.0 |
| speech-recognition-on-ami | microsoft/Phi-4-multimodal-instruct | – | Word Error Rate (WER): 10.69 |
| speech-recognition-on-ami-cleaned | microsoft/Phi-4-multimodal-instruct | – | Word Error Rate (WER): 9.11 |
| speech-recognition-on-earnings-22 | microsoft/Phi-4-multimodal-instruct | – | Word Error Rate (WER): 9.35 |
| speech-recognition-on-gigaspeech-cleaned | microsoft/Phi-4-multimodal-instruct | – | Word Error Rate (WER): 7.8 |
| speech-recognition-on-gigaspeech-test | microsoft/Phi-4-multimodal-instruct | – | Word Error Rate (WER): 7.91 |
| speech-recognition-on-librispeech-test-clean | microsoft/Phi-4-multimodal-instruct | – | Word Error Rate (WER): 1.35 |
| speech-recognition-on-librispeech-test-other | microsoft/Phi-4-multimodal-instruct | – | Word Error Rate (WER): 3.45 |
| speech-recognition-on-spgispeech | microsoft/Phi-4-multimodal-instruct | – | Word Error Rate (WER): 2.73 |
| speech-recognition-on-voxpopuli | microsoft/Phi-4-multimodal-instruct | – | Word Error Rate (WER): 5.77 |
| speech-recognition-on-voxpopuli-cleaned | microsoft/Phi-4-multimodal-instruct | – | Word Error Rate (WER): 4.18 |