Agentic Writing on DeepConsult
57.2Pass@1GPT-5 High
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5 High2026.03 | 57.2 | |
| Gemini 3-pro2026.03 | 48 | |
| Seed1.82026.03 | 41 | |
| Claude Sonnet-4.52026.03 | 38.3 | |
| Gemini 2.5-pro2026.03 | 21.8 |
Loading the SOTA2 catalog…
| Method | Links | |
|---|---|---|
| GPT-5 High2026.03 | 57.2 | |
| Gemini 3-pro2026.03 | 48 | |
| Seed1.82026.03 | 41 | |
| Claude Sonnet-4.52026.03 | 38.3 | |
| Gemini 2.5-pro2026.03 | 21.8 |