Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on GSM8K (Accuracy, Overall Average Accuracy)
Loading...
67.65
Accuracy
Standard FT
35.7324
44.0187
52.305
60.5913
Jun 1, 2026
Accuracy
Overall Average Accuracy
Updated 1d ago
Evaluation Results
Method
Method
Links
Accuracy
Overall Average Accuracy
Standard FT
Model=Gemma-3-4B
2026.06
67.65
55.93
AlphaToken
Model=Gemma-3-4B
2026.06
66.18
57.85
LESS
Model=Gemma-3-4B
2026.06
65.42
56.67
STM
Model=Gemma-3-4B
2026.06
64.15
56.24
ssTOKEN
Model=Gemma-3-4B
2026.06
63.6
55.82
Token Cleaning
Model=Gemma-3-4B
2026.06
63.2
55.57
LoRA
Model=Gemma-3-4B
2026.06
62.45
54.94
XTF
Model=Gemma-3-4B
2026.06
61.92
54.61
Pre-trained
Model=Gemma-3-4B
2026.06
36.96
43.96
Feedback
Search any
task
Search any
task