{"benchmark_id":"mmlongbench-128k","benchmark_name":"MMLongBench-128K","benchmark_description":"MMLongBench-128K evaluates multimodal long-context understanding at a 128K token context length, testing how well vision-language models reason over very long mixed text and image inputs.","max_score":1.0,"categories":["long_context","multimodal","reasoning","vision"],"modality":"multimodal","total_models":2,"entries":[{"rank":1,"model_id":"seed-2.1-pro","model_name":"Seed 2.1 Pro","organization_name":"ByteDance","organization_id":"bytedance","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":"2026-06-24","announcement_date":"2026-06-24","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"seed-2.1-turbo","model_name":"Seed 2.1 Turbo","organization_name":"ByteDance","organization_id":"bytedance","benchmark_score":0.769,"normalized_score":0.769,"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":"2026-06-24","announcement_date":"2026-06-24","multimodal":true,"param_count":null,"is_new":false}]}