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

Financial Open-ended QA on FinQA (test)

29.67Token Accuracy

DEEPFUSION

27.818828.299428.7829.2606Feb 15, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
29.67
2026.02
29.57
2026.02
29.47
2026.02
29.22
2026.02
29.06
2026.02
29.06
2026.02
29.04
2026.02
29.03
2026.02
28.95
2026.02
28.93
2026.02
28.88
2026.02
28.64
2026.02
28.32
2026.02
28.28
2026.02
28.26
2026.02
27.89