{"benchmark_id":"aa-lcr","benchmark_name":"AA-LCR","benchmark_description":"Agent Arena Long Context Reasoning benchmark","max_score":1.0,"categories":["long_context","reasoning"],"modality":"text","total_models":19,"entries":[{"rank":1,"model_id":"muse-glimmer-30b","model_name":"Muse Glimmer-30B","organization_name":"Meta","organization_id":"meta","benchmark_score":0.8,"normalized_score":0.8,"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":"2026-08-10","announcement_date":"2026-08-10","multimodal":true,"param_count":29600000000,"is_new":false},{"rank":2,"model_id":"hy3","model_name":"Hy3","organization_name":"Tencent","organization_id":"tencent","benchmark_score":0.734,"normalized_score":0.734,"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":"2026-07-06","announcement_date":"2026-07-06","multimodal":false,"param_count":295000000000,"is_new":false},{"rank":3,"model_id":"mistral-small-latest","model_name":"Mistral Small 4","organization_name":"Mistral AI","organization_id":"mistral","benchmark_score":0.712,"normalized_score":0.712,"verified":false,"self_reported":true,"provider_id":"mistral","input_cost_per_million":0.15,"output_cost_per_million":0.6,"speed_rps":null,"context_window":256000,"release_date":"2026-03-16","announcement_date":"2026-03-16","multimodal":true,"param_count":119000000000,"is_new":false},{"rank":4,"model_id":"solar-pro4","model_name":"Solar Pro 4","organization_name":"Upstage","organization_id":"upstage","benchmark_score":0.71,"normalized_score":0.71,"verified":false,"self_reported":true,"provider_id":"upstage","input_cost_per_million":0.3,"output_cost_per_million":1.2,"speed_rps":null,"context_window":524288,"release_date":"2026-08-06","announcement_date":"2026-08-06","multimodal":false,"param_count":null,"is_new":false},{"rank":5,"model_id":"kimi-k2.5","model_name":"Kimi K2.5","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.7,"normalized_score":0.7,"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":"2026-01-27","announcement_date":"2026-01-27","multimodal":true,"param_count":1000000000000,"is_new":false},{"rank":6,"model_id":"qwen3.5-397b-a17b","model_name":"Qwen3.5-397B-A17B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.687,"normalized_score":0.687,"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":"2026-02-16","announcement_date":"2026-02-16","multimodal":true,"param_count":397000000000,"is_new":false},{"rank":7,"model_id":"qwen3.6-plus","model_name":"Qwen3.6 Plus","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.683,"normalized_score":0.683,"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":8,"model_id":"qwen3.5-122b-a10b","model_name":"Qwen3.5-122B-A10B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.669,"normalized_score":0.669,"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":9,"model_id":"qwen3.5-27b","model_name":"Qwen3.5-27B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.661,"normalized_score":0.661,"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":10,"model_id":"nemotron-3-ultra-550b-a55b","model_name":"Nemotron 3 Ultra (550B A55B)","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.654,"normalized_score":0.654,"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":"2026-06-04","announcement_date":"2026-06-04","multimodal":false,"param_count":550000000000,"is_new":false},{"rank":11,"model_id":"qwen3.5-9b","model_name":"Qwen3.5-9B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.63,"normalized_score":0.63,"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":"2026-03-02","announcement_date":"2026-03-02","multimodal":true,"param_count":9000000000,"is_new":false},{"rank":12,"model_id":"minimax-m2.1","model_name":"MiniMax M2.1","organization_name":"MiniMax","organization_id":"minimax","benchmark_score":0.62,"normalized_score":0.62,"verified":false,"self_reported":true,"provider_id":"minimax","input_cost_per_million":0.3,"output_cost_per_million":1.2,"speed_rps":100.0,"context_window":1000000,"release_date":"2025-12-23","announcement_date":"2025-12-23","multimodal":false,"param_count":230000000000,"is_new":false},{"rank":13,"model_id":"qwen3.5-35b-a3b","model_name":"Qwen3.5-35B-A3B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.585,"normalized_score":0.585,"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":14,"model_id":"nemotron-3-super-120b-a12b","model_name":"Nemotron 3 Super (120B A12B)","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.5831,"normalized_score":0.5831,"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":"2026-03-11","announcement_date":"2026-03-11","multimodal":false,"param_count":120000000000,"is_new":false},{"rank":15,"model_id":"qwen3.5-4b","model_name":"Qwen3.5-4B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.57,"normalized_score":0.57,"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":"2026-03-02","announcement_date":"2026-03-02","multimodal":true,"param_count":4000000000,"is_new":false},{"rank":16,"model_id":"nemotron-3.5-lightning-30b-a3b","model_name":"Nemotron 3.5 Lightning (30B A3B)","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.52,"normalized_score":0.52,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.05,"output_cost_per_million":0.2,"speed_rps":null,"context_window":262144,"release_date":"2026-08-11","announcement_date":"2026-08-11","multimodal":false,"param_count":30000000000,"is_new":false},{"rank":17,"model_id":"exaone-4.5-33b","model_name":"EXAONE 4.5 33B","organization_name":"LG AI Research","organization_id":"lg","benchmark_score":0.506,"normalized_score":0.506,"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":"2026-04-09","announcement_date":"2026-04-09","multimodal":true,"param_count":33000000000,"is_new":false},{"rank":18,"model_id":"qwen3.5-2b","model_name":"Qwen3.5-2B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.256,"normalized_score":0.256,"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":"2026-03-02","announcement_date":"2026-03-02","multimodal":true,"param_count":2000000000,"is_new":false},{"rank":19,"model_id":"qwen3.5-0.8b","model_name":"Qwen3.5-0.8B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.047,"normalized_score":0.047,"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":"2026-03-02","announcement_date":"2026-03-02","multimodal":true,"param_count":800000000,"is_new":false}]}