Average of Benchmarks
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Average of 8 benchmarks
52.79Average Accuracy
72
Average of 7 benchmarks (including AIME24, LeetCode) (test)
52Average Accuracy
42
Average of 8 Benchmarks (ARC-C, ARC-E, BoolQ, HellaS, LamOp, Piqa, WinoG, MMLU)
72.99Accuracy
10
Average of 5 benchmarks (Spider-dev, Spider-DK, Spider-Realistic, Spider-test, Bird-dev)
70.2Accuracy
6