{"benchmark_id":"gorilla-benchmark-api-bench","benchmark_name":"Gorilla Benchmark API Bench","benchmark_description":"APIBench, a comprehensive dataset of over 11,000 instruction-API pairs from HuggingFace, TorchHub, and TensorHub APIs for evaluating language models' ability to generate accurate API calls.","max_score":1.0,"categories":["reasoning","code","tool_calling"],"modality":"text","total_models":3,"entries":[{"rank":1,"model_id":"llama-3.1-405b-instruct","model_name":"Llama 3.1 405B Instruct","organization_name":"Meta","organization_id":"meta","benchmark_score":0.353,"normalized_score":0.353,"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":"2024-07-23","announcement_date":"2024-07-23","multimodal":false,"param_count":405000000000,"is_new":false},{"rank":2,"model_id":"llama-3.1-70b-instruct","model_name":"Llama 3.1 70B Instruct","organization_name":"Meta","organization_id":"meta","benchmark_score":0.297,"normalized_score":0.297,"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":"2024-07-23","announcement_date":"2024-07-23","multimodal":false,"param_count":70000000000,"is_new":false},{"rank":3,"model_id":"llama-3.1-8b-instruct","model_name":"Llama 3.1 8B Instruct","organization_name":"Meta","organization_id":"meta","benchmark_score":0.082,"normalized_score":0.082,"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":"2024-07-23","announcement_date":"2024-07-23","multimodal":false,"param_count":8000000000,"is_new":false}]}