{"benchmark_id":"figqa","benchmark_name":"FigQA","benchmark_description":"FigQA is a multiple-choice benchmark on interpreting scientific figures from biology papers. It evaluates dual-use biological knowledge and multimodal reasoning relevant to bioweapons development.","max_score":1.0,"categories":["safety","healthcare","vision"],"modality":"multimodal","total_models":3,"entries":[{"rank":1,"model_id":"claude-mythos-preview","model_name":"Claude Mythos Preview","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.89,"normalized_score":0.89,"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":null,"announcement_date":"2026-04-07","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"claude-opus-4-6","model_name":"Claude Opus 4.6","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.783,"normalized_score":0.783,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":5.0,"output_cost_per_million":25.0,"speed_rps":42.0,"context_window":1000000,"release_date":"2026-02-05","announcement_date":"2026-02-05","multimodal":true,"param_count":null,"is_new":false},{"rank":3,"model_id":"grok-4.1-thinking-2025-11-17","model_name":"Grok-4.1 Thinking","organization_name":"xAI","organization_id":"xai","benchmark_score":0.34,"normalized_score":0.34,"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":"2025-11-17","announcement_date":"2025-11-17","multimodal":true,"param_count":null,"is_new":false}]}