{"benchmark_id":"labbench2","benchmark_name":"LABBench2","benchmark_description":"LABBench2 evaluates models on real-world biology research tasks.","max_score":1.0,"categories":["reasoning","science","agents","biology"],"modality":"text","total_models":3,"entries":[{"rank":1,"model_id":"gemini-4-argon","model_name":"Gemini 4 Argon","organization_name":"Google","organization_id":"google","benchmark_score":0.888,"normalized_score":0.888,"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":"2026-09-30","announcement_date":"2026-09-30","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"gemini-3.8-flash","model_name":"Gemini 3.8 Flash","organization_name":"Google","organization_id":"google","benchmark_score":0.862,"normalized_score":0.862,"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":"2026-09-02","announcement_date":"2026-09-02","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":0.821,"normalized_score":0.821,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.75,"output_cost_per_million":3.75,"speed_rps":null,"context_window":1000000,"release_date":"2026-08-13","announcement_date":"2026-08-13","multimodal":true,"param_count":null,"is_new":false}]}