World Knowledge on MMLU, MMLU-pro, SuperGPQA, LPFQA
93.8Pass@1 ScoreGPT-5 High
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-5 HighDataset=MMLU2026.03 | 93.8 | |
| Gemini 3-proDataset=MMLU2026.03 | 93.8 | |
| Claude Sonnet-4.5Dataset=MMLU2026.03 | 93.1 | |
| Gemini 2.5-proDataset=MMLU2026.03 | 92.9 | |
| Seed1.8Dataset=MMLU2026.03 | 92.3 | |
| Gemini 3-proDataset=MMLU-pro2026.03 | 90.2 | |
| Claude Sonnet-4.5Dataset=MMLU-pro2026.03 | 88.8 | |
| GPT-5 HighDataset=MMLU-pro2026.03 | 87.2 | |
| Gemini 2.5-proDataset=MMLU-pro2026.03 | 86.9 | |
| Seed1.8Dataset=MMLU-pro2026.03 | 84.9 | |
| Gemini 3-proDataset=SuperGPQA2026.03 | 75.3 | |
| GPT-5 HighDataset=SuperGPQA2026.03 | 66.8 | |
| Claude Sonnet-4.5Dataset=SuperGPQA2026.03 | 66.1 | |
| Gemini 2.5-proDataset=SuperGPQA2026.03 | 64.9 | |
| Seed1.8Dataset=SuperGPQA2026.03 | 64.8 | |
| GPT-5 HighDataset=LPFQA2026.03 | 54.4 | |
| Gemini 3-proDataset=LPFQA2026.03 | 51.1 | |
| Claude Sonnet-4.5Dataset=LPFQA2026.03 | 49.5 | |
| Seed1.8Dataset=LPFQA2026.03 | 49.1 | |
| Gemini 2.5-proDataset=LPFQA2026.03 | 47.7 |