GSM8K
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
GSM8K
95.2BA Accuracy @20
2
GSM8K
18.14EM (strict)
2
GSM8K
1.59Accuracy
2
GSM8K
76.4Accuracy (n=1k) on GSM8K
2
GSM8K
8.9Harmful Score (N=1000)
2
GSM8k
0.292Score
2
GSM8K
0.6Performance Ratio (PR)
2
GSM8K
97.19Accuracy
2
GSM8K
97.8Raw ASR
2
GSM8K (first 500 samples)
62.8Accuracy
2
GSM8K
50.5NFE
2
GSM8K Hard
13.9Det. Accuracy
2
GSM8K
3PTR Advancement Accuracy
2
GSM8K
50Base CA
2
GSM8K p
99.2Accuracy
2
GSM8K full (test)
94.77Accuracy
2
GSM8K
97.1Accuracy (%)
2
GSM8K
509.7Parameters (Full) (M)
2
GSM8K n=300 (test)
73.7Monotonicity Rate
2
GSM8K DE
55Comp. Acc.
2
GSM8K
89.84Accuracy
2
GSM8K (test)
80.6AUROC
2
GSM8K
160.51Output Throughput
2
Generalized GSM8K VeRA (test)
98.56Pass@1 Accuracy
2
GSM8K
93.37Score
2