{"benchmark_id":"longfact","benchmark_name":"LongFact","benchmark_description":"LongFact evaluates factual precision over long-form generations containing many individual claims. Each claim is extracted and verified, and the model is scored on claim-level precision, measuring whether extended responses introduce unsupported or false statements.","max_score":1.0,"categories":["factuality","general"],"modality":"text","total_models":1,"entries":[{"rank":1,"model_id":"mai-thinking-1","model_name":"MAI-Thinking-1","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.98,"normalized_score":0.98,"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":1000000000000,"is_new":false}]}