{"benchmark_id":"robust-if","benchmark_name":"Robust IF","benchmark_description":"Robust IF evaluates instruction-following robustness on diverse, hard prompts, measuring whether a model reliably adheres to constraints across challenging single-turn and multi-turn scenarios.","max_score":1.0,"categories":["reasoning","instruction_following"],"modality":"text","total_models":1,"entries":[{"rank":1,"model_id":"mai-code-1-flash","model_name":"MAI-Code-1-Flash","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.612,"normalized_score":0.612,"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}]}