Search Agent Capacity on HLE Verified
73.6ScoreSeed2.0 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Seed2.0 Pro2026.06 | 73.6 | |
| Gemini-3-Flash High2026.06 | 71.8 | |
| Seed2.0 Lite2026.06 | 70.7 | |
| GPT-5.2 High2026.06 | 68.5 | |
| Gemini-3-pro High2026.06 | 67.5 | |
| Claude-Opus-4.52026.06 | 56.6 | |
| GPT-5-mini High2026.06 | 56.4 | |
| Claude-Sonnet-4.52026.06 | 37.6 |