Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Mathematical Reasoning on GSM8K (100-question subset, test)

82Accuracy

SnapKV-D

31.0444.2757.570.73Dec 12, 2025
Updated 4d ago

Evaluation Results

MethodLinks
2025.12
82
2025.12
82
2025.12
81
2025.12
80
2025.12
80
2025.12
78
2025.12
70
2025.12
66
2025.12
64
2025.12
62
2025.12
56
2025.12
33