{"benchmark_id":"frontiermath","benchmark_name":"FrontierMath","benchmark_description":"A benchmark of hundreds of original, exceptionally challenging mathematics problems crafted and vetted by expert mathematicians, covering most major branches of modern mathematics from number theory and real analysis to algebraic geometry and category theory.","max_score":1.0,"categories":["math","reasoning"],"modality":"text","total_models":17,"entries":[{"rank":1,"model_id":"gpt-5.6-sol","model_name":"GPT-5.6 Sol","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.89,"normalized_score":0.89,"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":0.849,"normalized_score":0.849,"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":0.786,"normalized_score":0.786,"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},{"rank":4,"model_id":"gpt-5.4","model_name":"GPT-5.4","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.476,"normalized_score":0.476,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.5,"output_cost_per_million":15.0,"speed_rps":50.0,"context_window":1000000,"release_date":"2026-03-05","announcement_date":"2026-03-05","multimodal":true,"param_count":null,"is_new":false},{"rank":5,"model_id":"gpt-5.2-2025-12-11","model_name":"GPT-5.2","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.403,"normalized_score":0.403,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":1.75,"output_cost_per_million":14.0,"speed_rps":100.0,"context_window":400000,"release_date":"2025-12-11","announcement_date":"2025-12-11","multimodal":true,"param_count":null,"is_new":false},{"rank":6,"model_id":"gpt-5.5-pro","model_name":"GPT-5.5 Pro","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.396,"normalized_score":0.396,"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":"2026-04-23","announcement_date":"2026-04-23","multimodal":true,"param_count":null,"is_new":false},{"rank":7,"model_id":"gpt-5.5","model_name":"GPT-5.5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.354,"normalized_score":0.354,"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":"2026-04-23","announcement_date":"2026-04-23","multimodal":true,"param_count":null,"is_new":false},{"rank":8,"model_id":"gpt-5.1-2025-11-13","model_name":"GPT-5.1","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.267,"normalized_score":0.267,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":1.25,"output_cost_per_million":10.0,"speed_rps":100.0,"context_window":400000,"release_date":"2025-11-13","announcement_date":"2025-11-13","multimodal":true,"param_count":null,"is_new":false},{"rank":9,"model_id":"gpt-5.1-instant-2025-11-12","model_name":"GPT-5.1 Instant","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.267,"normalized_score":0.267,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":1.25,"output_cost_per_million":10.0,"speed_rps":100.0,"context_window":400000,"release_date":"2025-11-12","announcement_date":"2025-11-12","multimodal":true,"param_count":null,"is_new":false},{"rank":10,"model_id":"gpt-5.1-thinking-2025-11-12","model_name":"GPT-5.1 Thinking","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.267,"normalized_score":0.267,"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":"2025-11-12","announcement_date":"2025-11-12","multimodal":true,"param_count":null,"is_new":false},{"rank":11,"model_id":"gpt-5-2025-08-07","model_name":"GPT-5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.263,"normalized_score":0.263,"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":"2025-08-07","announcement_date":"2025-08-07","multimodal":true,"param_count":null,"is_new":false},{"rank":12,"model_id":"gpt-5-mini-2025-08-07","model_name":"GPT-5 mini","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.221,"normalized_score":0.221,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.25,"output_cost_per_million":2.0,"speed_rps":200.0,"context_window":400000,"release_date":"2025-08-07","announcement_date":"2025-08-07","multimodal":true,"param_count":null,"is_new":false},{"rank":13,"model_id":"o3-2025-04-16","model_name":"o3","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.158,"normalized_score":0.158,"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":"2025-04-16","announcement_date":"2025-04-16","multimodal":true,"param_count":null,"is_new":false},{"rank":14,"model_id":"gpt-5-nano-2025-08-07","model_name":"GPT-5 nano","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.096,"normalized_score":0.096,"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":"2025-08-07","announcement_date":"2025-08-07","multimodal":true,"param_count":null,"is_new":false},{"rank":15,"model_id":"o3-mini","model_name":"o3-mini","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.092,"normalized_score":0.092,"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":"2025-01-30","announcement_date":"2025-01-30","multimodal":false,"param_count":null,"is_new":false},{"rank":16,"model_id":"mai-code-1-flash","model_name":"MAI-Code-1-Flash","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.063,"normalized_score":0.063,"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":"2026-06-02","announcement_date":"2026-06-02","multimodal":false,"param_count":null,"is_new":false},{"rank":17,"model_id":"o1-2024-12-17","model_name":"o1","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.055,"normalized_score":0.055,"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":"2024-12-17","announcement_date":"2024-12-17","multimodal":false,"param_count":null,"is_new":false}]}