{"benchmark_id":"genebench-pro","benchmark_name":"GeneBench-Pro","benchmark_description":"GeneBench-Pro is a research-level benchmark of 129 multi-stage computational-biology problems spanning genomics, quantitative biology, and translational biomedicine. Each problem gives the agent a messy dataset, brief context, and a target estimand, and requires navigating dependent inferential decision points to reach a verifiable answer.","max_score":1.0,"categories":["reasoning","science","agents"],"modality":"text","total_models":3,"entries":[{"rank":1,"model_id":"gpt-5.6-sol","model_name":"GPT-5.6 Sol","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.287,"normalized_score":0.287,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":5.0,"output_cost_per_million":30.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"gpt-5.6-terra","model_name":"GPT-5.6 Terra","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.233,"normalized_score":0.233,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.0,"output_cost_per_million":12.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":3,"model_id":"gpt-5.6-luna","model_name":"GPT-5.6 Luna","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.108,"normalized_score":0.108,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.2,"output_cost_per_million":1.2,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false}]}