{"benchmark_id":"mlvu","benchmark_name":"MLVU","benchmark_description":"A comprehensive benchmark for multi-task long video understanding that evaluates multimodal large language models on videos ranging from 3 minutes to 2 hours across 9 distinct tasks including reasoning, captioning, recognition, and summarization.","max_score":1.0,"categories":["long_context","multimodal","video","vision"],"modality":"multimodal","total_models":10,"entries":[{"rank":1,"model_id":"qwen3.7-plus","model_name":"Qwen3.7-Plus","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.874,"normalized_score":0.874,"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":"2026-05-31","announcement_date":"2026-05-31","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"qwen3.5-122b-a10b","model_name":"Qwen3.5-122B-A10B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.873,"normalized_score":0.873,"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":"2026-02-24","announcement_date":"2026-02-24","multimodal":true,"param_count":122000000000,"is_new":false},{"rank":3,"model_id":"qwen3.6-plus","model_name":"Qwen3.6 Plus","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.867,"normalized_score":0.867,"verified":false,"self_reported":true,"provider_id":"together","input_cost_per_million":0.5,"output_cost_per_million":3.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-03-31","announcement_date":"2026-04-02","multimodal":true,"param_count":null,"is_new":false},{"rank":4,"model_id":"qwen3.6-27b","model_name":"Qwen3.6-27B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.866,"normalized_score":0.866,"verified":false,"self_reported":true,"provider_id":"novita","input_cost_per_million":0.6,"output_cost_per_million":3.6,"speed_rps":null,"context_window":262144,"release_date":"2026-04-21","announcement_date":"2026-04-21","multimodal":true,"param_count":27781427952,"is_new":false},{"rank":5,"model_id":"qwen3.6-35b-a3b","model_name":"Qwen3.6-35B-A3B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.862,"normalized_score":0.862,"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":"2026-04-16","announcement_date":"2026-04-16","multimodal":true,"param_count":35000000000,"is_new":false},{"rank":6,"model_id":"qwen3.5-27b","model_name":"Qwen3.5-27B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.859,"normalized_score":0.859,"verified":false,"self_reported":true,"provider_id":"novita","input_cost_per_million":0.3,"output_cost_per_million":2.4,"speed_rps":null,"context_window":262144,"release_date":"2026-02-24","announcement_date":"2026-02-24","multimodal":true,"param_count":27000000000,"is_new":false},{"rank":7,"model_id":"qwen3.5-35b-a3b","model_name":"Qwen3.5-35B-A3B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.856,"normalized_score":0.856,"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":"2026-02-24","announcement_date":"2026-02-24","multimodal":true,"param_count":35000000000,"is_new":false},{"rank":8,"model_id":"qwen3-vl-235b-a22b-instruct","model_name":"Qwen3 VL 235B A22B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.843,"normalized_score":0.843,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":236000000000,"is_new":false},{"rank":9,"model_id":"qwen3-vl-235b-a22b-thinking","model_name":"Qwen3 VL 235B A22B Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.838,"normalized_score":0.838,"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":"2025-09-22","announcement_date":"2025-09-22","multimodal":true,"param_count":236000000000,"is_new":false},{"rank":10,"model_id":"qwen2.5-vl-7b","model_name":"Qwen2.5 VL 7B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.702,"normalized_score":0.702,"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}]}