{"benchmark_id":"mmbench-video","benchmark_name":"MMBench-Video","benchmark_description":"A long-form multi-shot benchmark for holistic video understanding that incorporates approximately 600 web videos from YouTube spanning 16 major categories, with each video ranging from 30 seconds to 6 minutes. Includes roughly 2,000 original question-answer pairs covering 26 fine-grained capabilities.","max_score":1.0,"categories":["multimodal","reasoning","video","vision"],"modality":"multimodal","total_models":3,"entries":[{"rank":1,"model_id":"qwen2.5-vl-72b","model_name":"Qwen2.5 VL 72B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.0202,"normalized_score":0.0202,"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":"2025-01-26","announcement_date":"2025-01-26","multimodal":true,"param_count":72000000000,"is_new":false},{"rank":2,"model_id":"qwen2.5-vl-32b","model_name":"Qwen2.5 VL 32B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.0193,"normalized_score":0.0193,"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":"2025-02-28","announcement_date":"2025-02-28","multimodal":true,"param_count":33500000000,"is_new":false},{"rank":3,"model_id":"qwen2.5-vl-7b","model_name":"Qwen2.5 VL 7B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.0179,"normalized_score":0.0179,"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":"2025-01-26","announcement_date":"2025-01-26","multimodal":true,"param_count":8290000000,"is_new":false}]}