Scientific Knowledge and Reasoning on HLE no tool, text only
31.7ScoreGemini-3-Flash
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-3-FlashTier=High2026.06 | 31.7 | |
| Seed2.0 Lite2026.06 | 28.2 | |
| GPT-5-miniTier=High2026.06 | 17.6 | |
| Seed2.0 Mini2026.06 | 13.3 |
Loading the SOTA2 catalog…