{"benchmark_id":"longfact-objects","benchmark_name":"LongFact Objects","benchmark_description":"LongFact is a benchmark for evaluating long-form factuality in large language models. It comprises 2,280 fact-seeking prompts spanning 38 topics, designed to test a model's ability to generate accurate, long-form responses. The benchmark uses SAFE (Search-Augmented Factuality Evaluator) to evaluate factual accuracy.","max_score":1.0,"categories":["reasoning","general"],"modality":"text","total_models":1,"entries":[{"rank":1,"model_id":"gpt-5-2025-08-07","model_name":"GPT-5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.008,"normalized_score":0.008,"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}]}