| question-answering-on-drop-3-shot | DeepSeek-R1 (3-shot) | #1 | F1: 92.2 |
| question-answering-on-frames | DeepSeek-R1 | #1 | Accuracy: 82.5 |
| multi-task-language-understanding-on-c-eval | DeepSeek-R1-Zero | #2 | Accuracy: 92.8 |
| question-answering-on-simpleqa | DeepSeek-R1-Zero | #3 | Accuracy: 30.3 |
| code-generation-on-aider-polyglot | DeepSeek-R1 | #4 | Accuracy: 53.3 |
| question-answering-on-c-simpleqa | DeepSeek-R1-Zero | #4 | Accuracy: 66.4 |
| question-answering-on-newsqa | deepseek-r1 | #4 | EM: 80.57F1: 86.13 |
| code-generation-on-humaneval | DeepSeek R1 | #5 | Pass@1: 97.4 |
| instruction-following-on-ifeval | DeepSeek-R1 | #7 | Prompt-level strict-accuracy: 83.3 |
| code-generation-on-livecodebench-v5 | DeepSeek-R1 | #8 | Pass@1: 65.9 |
| multi-task-language-understanding-on-mmlu-redux | DeepSeek-R1 | #9 | Accuracy: 92.9 |
| mathematical-reasoning-on-math-500 | DeepSeek-R1 | #13 | Accuracy: 97.3 |
| multi-task-language-understanding-on-mmlu-pro | DeepSeek-R1 | #24 | Accuracy: 84.0 |
| mathematical-reasoning-on-aime24 | DeepSeek-r1 | #31 | Acc: 79.8 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | DeepSeek-R1-Zero | #66 | Accuracy: 75.8 |
| code-generation-on-swe-bench-verified | DeepSeek-R1 | #76 | Accuracy: 49.2 |