Reasoning on IMO-AnswerBench
89.8Avg@8DS-V4-Pro
Evaluation results
| Method | Links | |
|---|---|---|
| DS-V4-Pro2026.06 | 89.8 | |
| Claude Opus-4.72026.06 | 87.31 | |
| Ring-2.6-1Tvariant=xhigh2026.06 | 86.12 | |
| Kimi-K2.62026.06 | 86 | |
| Claude Opus-4.62026.06 | 80.28 | |
| OpenAI GPT-5.42026.06 | 77.53 | |
| Ring-2.6-1Tvariant=high2026.06 | 66.44 |