ResearchDatasetsGPQA, HumanEval, LiveCodeBench, AIME, MATHGPQA, HumanEval, LiveCodeBench, AIME, MATHFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsLarge Language Model Reasoning and CodingGPQA-Diamond, HumanEval, LiveCodeBench v6, AIME25, and MATH50088.3Throughput (Tokens/User)10
Large Language Model Reasoning and CodingGPQA-Diamond, HumanEval, LiveCodeBench v6, AIME25, and MATH50088.3Throughput (Tokens/User)10