Mathematical Problem Solving on IMOAnswerBench no tool
84.4ScoreGemini-3-Flash
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini-3-FlashTier=High2026.06 | 84.4 | |
| Seed2.0 Lite2026.06 | 81.6 | |
| GPT-5-miniTier=High2026.06 | 72.1 | |
| Seed2.0 Mini2026.06 | 71.6 |
| Method | Links | |
|---|---|---|
| Gemini-3-FlashTier=High2026.06 | 84.4 | |
| Seed2.0 Lite2026.06 | 81.6 | |
| GPT-5-miniTier=High2026.06 | 72.1 | |
| Seed2.0 Mini2026.06 | 71.6 |