Visual Puzzles on ZeroBench main
12Pass@1Seed2.0 Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Seed2.0 Pro2026.06 | 12 | |
| GPT-5.2 High2026.06 | 11 | |
| Seed1.82026.06 | 11 | |
| Gemini-3-Pro High2026.06 | 10 | |
| Seed2.0 Lite2026.06 | 8 | |
| Seed2.0 Mini2026.06 | 7 | |
| Claude-Opus-4.52026.06 | 4 |
Loading the SOTA2 catalog…