{"benchmark_id":"we-math","benchmark_name":"We-Math","benchmark_description":"We-Math evaluates multimodal models on visual mathematical reasoning, requiring models to understand and solve math problems presented with visual elements such as diagrams, charts, and geometric figures.","max_score":1.0,"categories":["math","reasoning","vision"],"modality":"multimodal","total_models":1,"entries":[{"rank":1,"model_id":"qwen3.6-plus","model_name":"Qwen3.6 Plus","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.89,"normalized_score":0.89,"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}]}