{"benchmark_id":"ds-fim-eval","benchmark_name":"DS-FIM-Eval","benchmark_description":"DeepSeek's internal Fill-in-the-Middle evaluation dataset for measuring code completion performance improvements in data science contexts","max_score":1.0,"categories":["general"],"modality":"text","total_models":1,"entries":[{"rank":1,"model_id":"deepseek-v2.5","model_name":"DeepSeek-V2.5","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.783,"normalized_score":0.783,"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":"2024-05-08","announcement_date":"2024-05-08","multimodal":false,"param_count":236000000000,"is_new":false}]}