Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Math Word Problem Solving on GSM8K (Acc, AUROC, ECE)
Loading...
0.9636
Accuracy
Qwen3 30B
0.641928
0.725439
0.80895
0.892461
Feb 18, 2026
Accuracy
AUROC
ECE
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
AUROC
ECE
Qwen3 30B
confidence_type=normal...
2026.02
0.9636
0.787
0.026
Gemma3 12B
confidence_type=normal...
2026.02
0.9287
0.801
0.045
Qwen3 4B
confidence_type=normal...
2026.02
0.9249
0.773
0.069
Gemma3 4B
confidence_type=normal...
2026.02
0.8605
0.686
0.121
GLM4 9B
confidence_type=normal...
2026.02
0.6543
0.734
0.137
Feedback
Search any
task
Search any
task