{"benchmark_id":"instruct-humaneval","benchmark_name":"Instruct HumanEval","benchmark_description":"Instruction-based variant of HumanEval benchmark for evaluating large language models' code generation capabilities with functional correctness using pass@k metric on programming problems","max_score":1.0,"categories":["general"],"modality":"text","total_models":1,"entries":[{"rank":1,"model_id":"llama-3.1-nemotron-70b-instruct","model_name":"Llama 3.1 Nemotron 70B Instruct","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.7384,"normalized_score":0.7384,"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":"2024-10-01","announcement_date":"2024-10-01","multimodal":false,"param_count":70000000000,"is_new":false}]}