GSM8K
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
GSM8K
3.91Solve Rate (SR)
16
GSM8K
87.9GSM Accuracy
16
GSM8K DeepSeek-distilled (test)
75.36Accuracy
16
GSM8K
1Average Rank
16
GSM8K Complex Samples (test)
26.1Avg@8 Score
16
GSM8K Simple Samples (test)
99.29Accuracy (avg@8)
16
GSM8K Insufficient
73.5Success Rate (SR)
16
GSM8K
48.64String Accuracy (Str.)
16
GSM8K (test)
19.57MAE
16
GSM8K (test)
92.3Accuracy
16
GSM8K
95.6Seed
16
GSM8K (test)
82.9Hit@1 Accuracy
16
GSM8K PI (Prompt Injection) (test)
3.3ASR@1
16
GSM8k (test)
1Accuracy
16
GSM8K
91.7Accuracy (GSM8K)
16
GSM8K
93.6Success Rate
16
GSM8K Aug
59.7Accuracy
15
GSM8K
86.68P@1
15
GSM8K
92Pass@1
15
GSM8K
1,828.7Throughput (tokens/s)
15
GSM8K
51.7Accuracy (ar)
15
GSM8K
97.71ASR
15
GSM8k
58.19Accuracy
15
GSM8K
4.44MAT Score (GSM8K)
15
GSM8K
91.37Accuracy
15