Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Spreadsheet Reasoning on SpreadsheetBench Verified
Loading...
71.11
Accuracy
SkillGrad
31.8188
42.0194
52.22
62.4206
May 26, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
SkillGrad
Backbone=GPT-5.4, Init...
2026.05
71.11
SkillGrad
Backbone=GPT-5.4, Init...
2026.05
69.44
EvoSkill
Backbone=GPT-5.4, Init...
2026.05
68.06
Trace2Skill
Backbone=GPT-5.4, Init...
2026.05
65.28
Trace2Skill
Backbone=GPT-5.4, Init...
2026.05
63.89
EvoSkill
Backbone=GPT-5.4, Init...
2026.05
63.61
No Skill
Backbone=GPT-5.4, Init...
2026.05
62.5
Base xlsx Skill
Backbone=GPT-5.4, Init...
2026.05
60
Base xlsx Skill
Backbone=GPT-5.4, Init...
2026.05
55.83
SkillGrad
Backbone=GPT-4.1, Init...
2026.05
54.17
SkillGrad
Backbone=GPT-4.1, Init...
2026.05
45.83
No Skill
Backbone=GPT-4.1, Init...
2026.05
44.17
Trace2Skill
Backbone=GPT-4.1, Init...
2026.05
38.89
Trace2Skill
Backbone=GPT-4.1, Init...
2026.05
37.22
EvoSkill
Backbone=GPT-4.1, Init...
2026.05
37.22
EvoSkill
Backbone=GPT-4.1, Init...
2026.05
36.94
Base xlsx Skill
Backbone=GPT-4.1, Init...
2026.05
36.67
Base xlsx Skill
Backbone=GPT-4.1, Init...
2026.05
33.33
Feedback
Search any
task
Search any
task