AMC
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
AMC23
91.9Avg@32
4
AMC23
99.9Avg@30 Accuracy
4
AMC 2023
86.3Pass@1 Success Rate (Last)
4
AMC23
8,732.8Average Length
4
AMC 23
86.9Accuracy (AMC 23)
4
AMC 23
74.53Pass@8
4
AMC
97.5Accuracy
4
AMC
47.06Mean@16
4
AMC23
65Accuracy
4
AMC (held-out)
85.9Accuracy
4
AMC23
89.2Pass@1
3
AMC23
19,376.5Average Output Length
3
AMC 23
39.69Mean@8 Accuracy
3
AMC23
13.34Pass@1 Accuracy
3
AMC 23
95.68pass@16
3
AMC
71.72Avg@32
3
AMC dataset
2.6Neural Cleanse Anomaly Index
3
AMC (out-of-distribution)
32.08Accuracy
3
AMC Out-of-distribution 2023 (test)
0.535Accuracy
3
AMC OpenR1-Math Harder
88.2Accuracy
3
AMC23
1,986.3TPS
3
AMC23
89.69Accuracy
2
AMC23
62.5P@1
2
AMC23
40Total Count
2
AMC
3.81Tau
2