All Benchmarks

Explore all 342 benchmarks for evaluating language models across different capabilities and domains

		Properties					Links
GPQA		text en	126	14	88.4%	57.3%
MMLU		text en	80	15	92.5%	79.8%
MMLU-Pro		text en	68	12	85.0%	65.6%
MATH		text en	64	11	97.9%	67.0%
HumanEval		text en	63	12	94.5%	80.6%
MMMU		multimodal en	52	11	84.2%	64.1%
LiveCodeBench		text en	48	10	80.4%	46.5%
GSM8k		text en	46	15	97.3%	87.8%
AIME 2024		text en	45	11	95.8%	73.2%
AIME 2025		text en	45	11	100.0%	67.6%

Showing 1 to 10 of 342 benchmarks

...