Track A Evaluation on 420-item set v4 (full)
51.9Exact Match Accuracygpt-5.2
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| gpt-5.22026.04 | 51.9 | 87.6 | |
| gpt-5.12026.04 | 48.1 | 85.2 | |
| glm-52026.04 | 47.1 | 86.2 | |
| deepseek-v3.12026.04 | 43.1 | 81.7 | |
| claude-opus-4.52026.04 | 41.9 | 85 | |
| Doubao-Seed-1.82026.04 | 38.6 | 75.7 | |
| glm-4.72026.04 | 37.6 | 76.4 | |
| claude-sonnet-4.52026.04 | 33.6 | 72.1 | |
| claude-haiku-4.52026.04 | 24.3 | 66.2 | |
| gemini-3-flash2026.04 | 9.8 | 52.6 | |
| MiniMax-M2.52026.04 | 0 | 24.3 | |
| MiniMax-M2.12026.04 | 0 | 18.8 |