{"benchmark_id":"graphwalks","benchmark_name":"GraphWalks","benchmark_description":"GraphWalks is a synthetic multi-hop long-context reasoning benchmark in which a model is given an edge-list representation of a graph and must traverse it to find neighboring nodes (via breadth-first search) or parent nodes for a given start node. Performance is reported as F1 of the model-predicted answer set versus the ground truth.","max_score":1.0,"categories":["long_context","reasoning"],"modality":"text","total_models":3,"entries":[{"rank":1,"model_id":"mai-thinking-1","model_name":"MAI-Thinking-1","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.9,"normalized_score":0.9,"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":"2026-06-02","announcement_date":"2026-06-02","multimodal":false,"param_count":1000000000000,"is_new":false},{"rank":2,"model_id":"mimo-v2.5","model_name":"MiMo-V2.5","organization_name":"Xiaomi","organization_id":"xiaomi","benchmark_score":0.87,"normalized_score":0.87,"verified":false,"self_reported":true,"provider_id":"novita","input_cost_per_million":0.168,"output_cost_per_million":0.336,"speed_rps":null,"context_window":1048576,"release_date":"2026-04-22","announcement_date":"2026-04-22","multimodal":true,"param_count":310775040000,"is_new":false},{"rank":3,"model_id":"mimo-v2.5-pro","model_name":"MiMo-V2.5-Pro","organization_name":"Xiaomi","organization_id":"xiaomi","benchmark_score":0.62,"normalized_score":0.62,"verified":false,"self_reported":true,"provider_id":"xiaomi","input_cost_per_million":0.435,"output_cost_per_million":0.87,"speed_rps":null,"context_window":1048576,"release_date":"2026-04-27","announcement_date":"2026-04-27","multimodal":false,"param_count":1023244718976,"is_new":false}]}