| MMLU | Llama-3.3-70b-Instruct | Accuracy82.4 | | 39 | 2mo ago |
| MMLU, MMLU-pro, SuperGPQA, LPFQA | | Pass@1 Score93.8 | | 20 | 4mo ago |
| JARVIS-VLA Benchmark 1.0 (test) | GPT-4o | Accuracy96.6 | | 10 | 4mo ago |
| MMLU | DeepSeek-V4-Pro Base | EM (World Knowledge)90.1 | | 7 | 1mo ago |
| MMLU-Pro | DeepSeek-V4-Pro Base | EM (World Knowledge)73.5 | | 6 | 1mo ago |
| KVFundaBench WK | | Accuracy72.99 | | 5 | 2mo ago |
| HUMANITY’S LAST EXAM text-only | | Score11.1 | | 4 | 4mo ago |
| GPQA Diamond | | Score77.5 | | 4 | 4mo ago |
| MMLU-PRO | DeepSeek-V3.2 | Score84.6 | | 4 | 4mo ago |
| TriviaQA | DeepSeek-V4-Pro Base | Exact Match (EM)85.6 | | 3 | 1mo ago |
| FACTS Parametric | DeepSeek-V4-Pro Base | EM62.6 | | 3 | 1mo ago |
| SuperGPQA | DeepSeek-V4-Pro Base | EM53.9 | | 3 | 1mo ago |
| Simple-QA verified | DeepSeek-V4-Pro Base | Exact Match (EM)55.2 | | 3 | 1mo ago |
| MultiLoKo | DeepSeek-V4-Pro Base | EM51.1 | | 3 | 1mo ago |
| CMMLU | DeepSeek-V4-Pro Base | Exact Match (EM)90.8 | | 3 | 1mo ago |
| C-Eval | DeepSeek-V4-Pro Base | EM93.1 | | 3 | 1mo ago |
| MMMLU | DeepSeek-V4-Pro Base | EM (Exact Match)90.3 | | 3 | 1mo ago |
| MMLU-Redux | DeepSeek-V4-Pro Base | Exact Match (EM)90.8 | | 3 | 1mo ago |
| AGIEval | DeepSeek-V4-Pro Base | EM83.1 | | 3 | 1mo ago |
| ARC-c | SYNPRO | Accuracy (ARC-c)41.47 | | 1 | 2mo ago |
| ARC-e | SYNPRO | Accuracy70.18 | | 1 | 2mo ago |