{"benchmark_id":"tau2-telecom","benchmark_name":"Tau2 Telecom","benchmark_description":"τ²-Bench telecom domain evaluates conversational agents in a dual-control environment modeled as a Dec-POMDP, where both agent and user use tools in shared telecommunications troubleshooting scenarios that test coordination and communication capabilities.","max_score":1.0,"categories":["reasoning","communication","tool_calling"],"modality":"text","total_models":36,"entries":[{"rank":1,"model_id":"claude-opus-4-6","model_name":"Claude Opus 4.6","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.993,"normalized_score":0.993,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":5.0,"output_cost_per_million":25.0,"speed_rps":42.0,"context_window":1000000,"release_date":"2026-02-05","announcement_date":"2026-02-05","multimodal":true,"param_count":null,"is_new":false},{"rank":2,"model_id":"longcat-flash-thinking-2601","model_name":"LongCat-Flash-Thinking-2601","organization_name":"Meituan","organization_id":"meituan","benchmark_score":0.993,"normalized_score":0.993,"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":"2026-01-14","announcement_date":"2026-01-14","multimodal":false,"param_count":560000000000,"is_new":false},{"rank":3,"model_id":"gpt-5.4","model_name":"GPT-5.4","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.989,"normalized_score":0.989,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.5,"output_cost_per_million":15.0,"speed_rps":50.0,"context_window":1000000,"release_date":"2026-03-05","announcement_date":"2026-03-05","multimodal":true,"param_count":null,"is_new":false},{"rank":4,"model_id":"gpt-5.2-2025-12-11","model_name":"GPT-5.2","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.987,"normalized_score":0.987,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":1.75,"output_cost_per_million":14.0,"speed_rps":100.0,"context_window":400000,"release_date":"2025-12-11","announcement_date":"2025-12-11","multimodal":true,"param_count":null,"is_new":false},{"rank":5,"model_id":"claude-opus-4-5-20251101","model_name":"Claude Opus 4.5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.982,"normalized_score":0.982,"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":"2025-11-24","announcement_date":"2025-11-24","multimodal":true,"param_count":null,"is_new":false},{"rank":6,"model_id":"gpt-5.5","model_name":"GPT-5.5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.98,"normalized_score":0.98,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":5.0,"output_cost_per_million":30.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-04-23","announcement_date":"2026-04-23","multimodal":true,"param_count":null,"is_new":false},{"rank":7,"model_id":"claude-sonnet-4-6","model_name":"Claude Sonnet 4.6","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.979,"normalized_score":0.979,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":3.0,"output_cost_per_million":15.0,"speed_rps":42.0,"context_window":200000,"release_date":"2026-02-17","announcement_date":"2026-02-17","multimodal":true,"param_count":null,"is_new":false},{"rank":8,"model_id":"mimo-v2-pro","model_name":"MiMo-V2-Pro","organization_name":"Xiaomi","organization_id":"xiaomi","benchmark_score":0.968,"normalized_score":0.968,"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":"2026-03-18","announcement_date":"2026-03-18","multimodal":false,"param_count":1000000000000,"is_new":false},{"rank":9,"model_id":"gpt-5-2025-08-07","model_name":"GPT-5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.967,"normalized_score":0.967,"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":"2025-08-07","announcement_date":"2025-08-07","multimodal":true,"param_count":null,"is_new":false},{"rank":10,"model_id":"gpt-5.1-2025-11-13","model_name":"GPT-5.1","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.956,"normalized_score":0.956,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":1.25,"output_cost_per_million":10.0,"speed_rps":100.0,"context_window":400000,"release_date":"2025-11-13","announcement_date":"2025-11-13","multimodal":true,"param_count":null,"is_new":false},{"rank":11,"model_id":"gpt-5.1-instant-2025-11-12","model_name":"GPT-5.1 Instant","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.956,"normalized_score":0.956,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":1.25,"output_cost_per_million":10.0,"speed_rps":100.0,"context_window":400000,"release_date":"2025-11-12","announcement_date":"2025-11-12","multimodal":true,"param_count":null,"is_new":false},{"rank":12,"model_id":"gpt-5.1-thinking-2025-11-12","model_name":"GPT-5.1 Thinking","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.956,"normalized_score":0.956,"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":"2025-11-12","announcement_date":"2025-11-12","multimodal":true,"param_count":null,"is_new":false},{"rank":13,"model_id":"gpt-5.4-mini","model_name":"GPT-5.4 mini","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.934,"normalized_score":0.934,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.75,"output_cost_per_million":4.5,"speed_rps":null,"context_window":400000,"release_date":"2026-03-17","announcement_date":"2026-03-17","multimodal":true,"param_count":null,"is_new":false},{"rank":14,"model_id":"nova-2-pro","model_name":"Nova 2 Pro","organization_name":"Amazon","organization_id":"amazon","benchmark_score":0.927,"normalized_score":0.927,"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":"2025-12-02","announcement_date":"2025-12-02","multimodal":true,"param_count":null,"is_new":false},{"rank":15,"model_id":"gpt-5.4-nano","model_name":"GPT-5.4 nano","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.925,"normalized_score":0.925,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.2,"output_cost_per_million":1.25,"speed_rps":null,"context_window":400000,"release_date":"2026-03-17","announcement_date":"2026-03-17","multimodal":true,"param_count":null,"is_new":false},{"rank":16,"model_id":"muse-spark","model_name":"Muse Spark","organization_name":"Meta","organization_id":"meta","benchmark_score":0.915,"normalized_score":0.915,"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":"2026-04-08","announcement_date":"2026-04-08","multimodal":true,"param_count":null,"is_new":false},{"rank":17,"model_id":"minimax-m2.1","model_name":"MiniMax M2.1","organization_name":"MiniMax","organization_id":"minimax","benchmark_score":0.87,"normalized_score":0.87,"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":18,"model_id":"minimax-m2","model_name":"MiniMax M2","organization_name":"MiniMax","organization_id":"minimax","benchmark_score":0.87,"normalized_score":0.87,"verified":false,"self_reported":true,"provider_id":"novita","input_cost_per_million":0.3,"output_cost_per_million":1.2,"speed_rps":null,"context_window":204800,"release_date":"2025-10-27","announcement_date":"2025-10-27","multimodal":false,"param_count":230000000000,"is_new":false},{"rank":19,"model_id":"command-a-plus-05-2026","model_name":"Command A+","organization_name":"Cohere","organization_id":"cohere","benchmark_score":0.85,"normalized_score":0.85,"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":"2026-05-20","announcement_date":"2026-05-20","multimodal":true,"param_count":218000000000,"is_new":false},{"rank":20,"model_id":"longcat-flash-thinking","model_name":"LongCat-Flash-Thinking","organization_name":"Meituan","organization_id":"meituan","benchmark_score":0.831,"normalized_score":0.831,"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":false,"param_count":560000000000,"is_new":false},{"rank":21,"model_id":"claude-haiku-4-5-20251001","model_name":"Claude Haiku 4.5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.83,"normalized_score":0.83,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":1.0,"output_cost_per_million":5.0,"speed_rps":100.0,"context_window":200000,"release_date":"2025-10-15","announcement_date":"2025-10-15","multimodal":true,"param_count":null,"is_new":false},{"rank":22,"model_id":"nova-2-omni","model_name":"Nova 2 Omni","organization_name":"Amazon","organization_id":"amazon","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":"2025-12-02","announcement_date":"2025-12-02","multimodal":true,"param_count":null,"is_new":false},{"rank":23,"model_id":"nova-2-lite","model_name":"Nova 2 Lite","organization_name":"Amazon","organization_id":"amazon","benchmark_score":0.76,"normalized_score":0.76,"verified":false,"self_reported":true,"provider_id":"bedrock","input_cost_per_million":0.3,"output_cost_per_million":2.5,"speed_rps":null,"context_window":1000000,"release_date":"2025-12-02","announcement_date":"2025-12-02","multimodal":true,"param_count":null,"is_new":false},{"rank":24,"model_id":"longcat-flash-chat","model_name":"LongCat-Flash-Chat","organization_name":"Meituan","organization_id":"meituan","benchmark_score":0.7368,"normalized_score":0.7368,"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":"2025-08-29","announcement_date":"2025-08-29","multimodal":false,"param_count":560000000000,"is_new":false},{"rank":25,"model_id":"exaone-4.5-33b","model_name":"EXAONE 4.5 33B","organization_name":"LG AI Research","organization_id":"lg","benchmark_score":0.73,"normalized_score":0.73,"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":26,"model_id":"longcat-flash-lite","model_name":"LongCat-Flash-Lite","organization_name":"Meituan","organization_id":"meituan","benchmark_score":0.728,"normalized_score":0.728,"verified":false,"self_reported":true,"provider_id":"meituan","input_cost_per_million":0.1,"output_cost_per_million":0.4,"speed_rps":500.0,"context_window":256000,"release_date":"2026-02-05","announcement_date":"2026-02-05","multimodal":false,"param_count":68500000000,"is_new":false},{"rank":27,"model_id":"mai-code-1-flash","model_name":"MAI-Code-1-Flash","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.717,"normalized_score":0.717,"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":null,"is_new":false},{"rank":28,"model_id":"kimi-k2-instruct-0905","model_name":"Kimi K2-Instruct-0905","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.658,"normalized_score":0.658,"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":"2025-09-05","announcement_date":"2025-09-05","multimodal":false,"param_count":1000000000000,"is_new":false},{"rank":29,"model_id":"kimi-k2-instruct","model_name":"Kimi K2 Instruct","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.658,"normalized_score":0.658,"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":"2025-07-11","announcement_date":"2025-07-11","multimodal":false,"param_count":1000000000000,"is_new":false},{"rank":30,"model_id":"nemotron-3-super-120b-a12b","model_name":"Nemotron 3 Super (120B A12B)","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.6436,"normalized_score":0.6436,"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":31,"model_id":"o3-2025-04-16","model_name":"o3","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.582,"normalized_score":0.582,"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":"2025-04-16","announcement_date":"2025-04-16","multimodal":true,"param_count":null,"is_new":false},{"rank":32,"model_id":"qwen3-235b-a22b-thinking-2507","model_name":"Qwen3-235B-A22B-Thinking-2507","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.456,"normalized_score":0.456,"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":"2025-07-25","announcement_date":"2025-07-25","multimodal":false,"param_count":235000000000,"is_new":false},{"rank":33,"model_id":"qwen3-next-80b-a3b-thinking","model_name":"Qwen3-Next-80B-A3B-Thinking","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.439,"normalized_score":0.439,"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":"2025-09-10","announcement_date":"2025-09-10","multimodal":false,"param_count":80000000000,"is_new":false},{"rank":34,"model_id":"nemotron-3-nano-30b-a3b","model_name":"Nemotron 3 Nano (30B A3B)","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.422,"normalized_score":0.422,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.06,"output_cost_per_million":0.24,"speed_rps":null,"context_window":262144,"release_date":"2025-12-15","announcement_date":"2025-12-15","multimodal":false,"param_count":32000000000,"is_new":false},{"rank":35,"model_id":"gpt-4o-2024-08-06","model_name":"GPT-4o","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.235,"normalized_score":0.235,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.5,"output_cost_per_million":10.0,"speed_rps":132.0,"context_window":128000,"release_date":"2024-08-06","announcement_date":"2024-08-06","multimodal":true,"param_count":null,"is_new":false},{"rank":36,"model_id":"qwen3-next-80b-a3b-instruct","model_name":"Qwen3-Next-80B-A3B-Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.132,"normalized_score":0.132,"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":"2025-09-10","announcement_date":"2025-09-10","multimodal":false,"param_count":80000000000,"is_new":false}]}