{"benchmark_id":"biomysterybench","benchmark_name":"BioMysteryBench","benchmark_description":"BioMysteryBench evaluates a model's ability to reason through challenging molecular biology problems, reporting performance on a hard subset and on the subset of problems solved by human experts.","max_score":1.0,"categories":["reasoning","science","biology"],"modality":"text","total_models":3,"entries":[{"rank":1,"model_id":"claude-opus-5","model_name":"Claude Opus 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.901,"normalized_score":0.901,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":5.0,"output_cost_per_million":25.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-07-24","announcement_date":"2026-07-24","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"claude-fable-5","model_name":"Claude Fable 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.461,"normalized_score":0.461,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":10.0,"output_cost_per_million":50.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-06-09","announcement_date":"2026-06-09","multimodal":true,"param_count":null,"is_new":false},{"rank":3,"model_id":"gemini-3.7-flash","model_name":"Gemini 3.7 Flash","organization_name":"Google","organization_id":"google","benchmark_score":0.435,"normalized_score":0.435,"verified":false,"self_reported":true,"provider_id":"google","input_cost_per_million":0.75,"output_cost_per_million":3.75,"speed_rps":null,"context_window":1048576,"release_date":"2026-08-13","announcement_date":"2026-08-13","multimodal":true,"param_count":null,"is_new":false}]}