mlfoundations-dev/

s1-32B_1744017028_eval_0981

Precomputed model outputs for evaluation.

	Evaluation Results




	
	Summary




	

Metric AIME24 AIME25 AMC23 MATH500 GPQADiamond LiveCodeBench

Accuracy 37.3 23.3 77.0 86.8 48.7 40.7

Average Accuracy: 37.33% ± 1.46% Number of Runs: 5

Run Accuracy Questions Solved Total Questions

1 36.67% 11 30

2 43.33% 13 30

3 36.67% 11 30

4 36.67% 11 30

5 33.33% 10 30