Science
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Science Benchmark
59.53Execution Accuracy
48
Science
79.4F1 Score
19
Science
0.028Cost ($)
15
Science single-parent hierarchies (test)
46.1R@1
13
Science
0.928Ranking Loss
11
Science
12.39Macro F1 Score
11
Science
25.122CV Score
11
Science (test)
3.44HL
11
Science
96OE Score
11
Science
5.26AP
11
Science (SCI) SemEval-2016 Task 13
13.2Chi-Squared
10
SCIENCE
76.37Accuracy@8
9
Science
30.5Perplexity
6
Science GPQA Diamond HLE (test)
63.1GPQA Diamond Score
6
Science (out-of-distribution)
65.12Accuracy
6
Science Curated Task Benchmark 1.0 (test)
0.005Average Cost
3
Science
44.7Prec@1
3
Science English
62.29F1 Score
2