Math Reasoning Evaluation Suite
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Math Reasoning Evaluation Suite (Math-500, AMC23, GSM8k, Minerva, Olympiad)
71.8Math-500
34
Math Reasoning Evaluation Suite (GSM8K, MATH, AMC23, OlympiadBench, GAOKAO-2024, Minerva, AIME) various (test)
95.8GSM8K Accuracy
7
Math-Reasoning Evaluation Suite (NumGLUE, MMLU-Math, GSM8K, SVAMP, SimulEq, DeepMind, AQuA, SAT)
57.68NumGLUE Accuracy
2