{"benchmark_id":"mme","benchmark_name":"MME","benchmark_description":"A comprehensive evaluation benchmark for Multimodal Large Language Models measuring both perception and cognition abilities across 14 subtasks. Features manually designed instruction-answer pairs to avoid data leakage and provides systematic quantitative assessment of MLLM capabilities.","max_score":1.0,"categories":["multimodal","reasoning","vision"],"modality":"multimodal","total_models":4,"entries":[{"rank":1,"model_id":"lfm-2.5-vl-3b","model_name":"LFM2.5-VL-3B","organization_name":"Liquid AI","organization_id":"liquid","benchmark_score":0.731,"normalized_score":0.731,"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":"2026-08-12","announcement_date":"2026-08-12","multimodal":true,"param_count":3123483888,"is_new":false},{"rank":2,"model_id":"deepseek-vl2","model_name":"DeepSeek VL2","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.2253,"normalized_score":0.2253,"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":"2024-12-13","announcement_date":"2024-12-13","multimodal":true,"param_count":27000000000,"is_new":false},{"rank":3,"model_id":"deepseek-vl2-small","model_name":"DeepSeek VL2 Small","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.2123,"normalized_score":0.2123,"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":"2024-12-13","announcement_date":"2024-12-13","multimodal":true,"param_count":16000000000,"is_new":false},{"rank":4,"model_id":"deepseek-vl2-tiny","model_name":"DeepSeek VL2 Tiny","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.1915,"normalized_score":0.1915,"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":"2024-12-13","announcement_date":"2024-12-13","multimodal":true,"param_count":3000000000,"is_new":false}]}