mlfoundations-dev/

verified_stratos_mix_no_proofs_without_metadata_eval_5554

Precomputed model outputs for evaluation.

	Evaluation Results




	
	Summary




	

Metric AIME24 AMC23 MATH500 MMLUPro JEEBench GPQADiamond LiveCodeBench CodeElo CodeForces HLE HMMT AIME25 LiveCodeBenchv5

Accuracy 30.3 67.5 83.4 53.5 45.2 43.3 19.1 11.3 15.3 11.4 13.3 26.7 27.5

Average Accuracy: 30.33% ± 1.10% Number of Runs: 10

Run Accuracy Questions… See the full description on the dataset page: https://huggingface.co/datasets/mlfoundations-dev/verified_stratos_mix_no_proofs_without_metadata_eval_5554.