General Reasoning on KOR-Bench ARC-AGI-1 (Pass@1)
77.4Pass@1GPT-5 High
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-5 HighDataset=KOR-Bench2026.03 | 77.4 | |
| Seed1.8Dataset=KOR-Bench2026.03 | 76.2 | |
| Gemini 3-proDataset=KOR-Bench2026.03 | 75 | |
| Gemini 3-proDataset=ARC-AGI-12026.03 | 75 | |
| Claude Sonnet-4.5Dataset=KOR-Bench2026.03 | 74.5 | |
| Gemini 2.5-proDataset=KOR-Bench2026.03 | 74.2 | |
| Seed1.8Dataset=ARC-AGI-12026.03 | 67.9 | |
| GPT-5 HighDataset=ARC-AGI-12026.03 | 65.7 | |
| Claude Sonnet-4.5Dataset=ARC-AGI-12026.03 | 63.7 | |
| Gemini 2.5-proDataset=ARC-AGI-12026.03 | 37 |