{"benchmark_id":"mmlu-redux-2.0","benchmark_name":"MMLU-redux-2.0","benchmark_description":"A curated version of the MMLU benchmark featuring manually re-annotated 5,700 questions across 57 subjects to identify and correct errors in the original dataset. Addresses the 6.49% error rate found in MMLU and provides more reliable evaluation metrics for language models.","max_score":1.0,"categories":["language","math","reasoning","general"],"modality":"text","total_models":1,"entries":[{"rank":1,"model_id":"kimi-k2-base","model_name":"Kimi K2 Base","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.902,"normalized_score":0.902,"verified":false,"self_reported":true,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"speed_rps":null,"context_window":null,"release_date":"2025-07-11","announcement_date":"2025-07-11","multimodal":false,"param_count":1000000000000,"is_new":false}]}