Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Spreadsheet Task Execution on SpreadsheetBench Verified (Held-out)

70.5Verification Score

SkillCAT

6.5423.14539.7556.355Jun 11, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
70.5
2026.06
69.83
2026.06
69.5
2026.06
64.5
2026.06
61
2026.06
55.5
2026.06
54.5
2026.06
48.33
2026.06
46
2026.06
46
2026.06
39
2026.06
39
2026.06
37
2026.06
32.5
2026.06
31
2026.06
31
2026.06
29.67
2026.06
26.33
2026.06
26.17
2026.06
20.17
2026.06
20
2026.06
9.67
2026.06
9
2026.06
9