Agentic Search on WideSearch
65.7Pass@1Claude Sonnet-4.5
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude Sonnet-4.52026.03 | 65.7 | |
| Seed1.82026.03 | 63.8 | |
| GPT-5 High2026.03 | 62.2 | |
| Gemini 3-pro2026.03 | 57 | |
| Gemini 2.5-pro2026.03 | 52.6 |
| Method | Links | |
|---|---|---|
| Claude Sonnet-4.52026.03 | 65.7 | |
| Seed1.82026.03 | 63.8 | |
| GPT-5 High2026.03 | 62.2 | |
| Gemini 3-pro2026.03 | 57 | |
| Gemini 2.5-pro2026.03 | 52.6 |