Spreadsheet Manipulation and Reasoning on SpreadsheetBench
28.8Soft Accuracy
TabClaw
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.06 | 28.8 | 24.5 | |
2026.06 | 24.1 | 18.9 | |
2026.06 | 23.3 | 19.8 | |
2026.06 | 22.3 | 18.5 | |
2026.06 | 18.8 | 15.5 | |
2026.06 | 18.8 | 16.1 | |
2026.06 | 14.8 | 12.9 | |
2026.06 | 14.8 | 12.2 |