{"benchmark_id":"apex-swe","benchmark_name":"APEX-SWE","benchmark_description":"APEX-SWE evaluates AI agents on software engineering tasks requiring multi-step coding, debugging, and verification.","max_score":1.0,"categories":["agents","code"],"modality":"text","total_models":1,"entries":[{"rank":1,"model_id":"grok-4.6","model_name":"Grok 4.6","organization_name":"xAI","organization_id":"xai","benchmark_score":0.564,"normalized_score":0.564,"verified":false,"self_reported":true,"provider_id":"xai","input_cost_per_million":2.0,"output_cost_per_million":6.0,"speed_rps":null,"context_window":500000,"release_date":"2026-08-12","announcement_date":"2026-08-12","multimodal":true,"param_count":null,"is_new":false}]}