{"benchmark_id":"terminal-bench-2","benchmark_name":"Terminal-Bench 2.0","benchmark_description":"Terminal-Bench 2.0 is an updated benchmark for testing AI agents' tool use ability to operate a computer via terminal. It evaluates how well models can handle real-world, end-to-end tasks autonomously, including compiling code, training models, setting up servers, system administration, security tasks, data science workflows, and cybersecurity vulnerabilities.","max_score":1.0,"categories":["reasoning","agents","code","tool_calling"],"modality":"text","total_models":51,"entries":[{"rank":1,"model_id":"gpt-5.5","model_name":"GPT-5.5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.827,"normalized_score":0.827,"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":2,"model_id":"claude-mythos-preview","model_name":"Claude Mythos Preview","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.82,"normalized_score":0.82,"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":null,"announcement_date":"2026-04-07","multimodal":true,"param_count":null,"is_new":false},{"rank":3,"model_id":"claude-sonnet-5","model_name":"Claude Sonnet 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.804,"normalized_score":0.804,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":2.0,"output_cost_per_million":10.0,"speed_rps":42.0,"context_window":1000000,"release_date":"2026-06-30","announcement_date":"2026-06-30","multimodal":true,"param_count":null,"is_new":false},{"rank":4,"model_id":"gpt-5.3-codex","model_name":"GPT-5.3 Codex","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.773,"normalized_score":0.773,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":1.75,"output_cost_per_million":14.0,"speed_rps":50.0,"context_window":400000,"release_date":"2026-02-05","announcement_date":"2026-02-05","multimodal":true,"param_count":null,"is_new":false},{"rank":5,"model_id":"gemini-3.5-flash","model_name":"Gemini 3.5 Flash","organization_name":"Google","organization_id":"google","benchmark_score":0.762,"normalized_score":0.762,"verified":false,"self_reported":true,"provider_id":"google","input_cost_per_million":1.5,"output_cost_per_million":9.0,"speed_rps":null,"context_window":1048576,"release_date":"2026-05-19","announcement_date":"2026-05-19","multimodal":true,"param_count":null,"is_new":false},{"rank":6,"model_id":"gpt-5.4","model_name":"GPT-5.4","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.751,"normalized_score":0.751,"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":7,"model_id":"claude-opus-4-8","model_name":"Claude Opus 4.8","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.746,"normalized_score":0.746,"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":"2026-05-28","announcement_date":"2026-05-28","multimodal":true,"param_count":null,"is_new":false},{"rank":8,"model_id":"qwen3.7-plus","model_name":"Qwen3.7-Plus","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.703,"normalized_score":0.703,"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":"2026-05-31","announcement_date":"2026-05-31","multimodal":true,"param_count":null,"is_new":false},{"rank":9,"model_id":"qwen3.7-max","model_name":"Qwen3.7 Max","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.697,"normalized_score":0.697,"verified":false,"self_reported":true,"provider_id":"novita","input_cost_per_million":1.25,"output_cost_per_million":3.75,"speed_rps":null,"context_window":1000000,"release_date":"2026-05-19","announcement_date":"2026-05-19","multimodal":false,"param_count":null,"is_new":false},{"rank":10,"model_id":"claude-opus-4-7","model_name":"Claude Opus 4.7","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.694,"normalized_score":0.694,"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":"2026-04-16","announcement_date":"2026-04-16","multimodal":true,"param_count":null,"is_new":false},{"rank":11,"model_id":"glm-5.1","model_name":"GLM-5.1","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.69,"normalized_score":0.69,"verified":false,"self_reported":true,"provider_id":"friendli","input_cost_per_million":1.4,"output_cost_per_million":4.4,"speed_rps":null,"context_window":200000,"release_date":"2026-04-07","announcement_date":"2026-04-07","multimodal":false,"param_count":754000000000,"is_new":false},{"rank":12,"model_id":"gemini-3.1-pro-preview","model_name":"Gemini 3.1 Pro","organization_name":"Google","organization_id":"google","benchmark_score":0.685,"normalized_score":0.685,"verified":false,"self_reported":true,"provider_id":"google","input_cost_per_million":2.5,"output_cost_per_million":15.0,"speed_rps":90.0,"context_window":1048576,"release_date":"2026-02-19","announcement_date":"2026-02-19","multimodal":true,"param_count":null,"is_new":false},{"rank":13,"model_id":"mimo-v2.5-pro","model_name":"MiMo-V2.5-Pro","organization_name":"Xiaomi","organization_id":"xiaomi","benchmark_score":0.684,"normalized_score":0.684,"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},{"rank":14,"model_id":"deepseek-v4-pro-max","model_name":"DeepSeek-V4-Pro-Max","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.679,"normalized_score":0.679,"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":"2026-04-23","announcement_date":"2026-04-23","multimodal":false,"param_count":1600000000000,"is_new":false},{"rank":15,"model_id":"kimi-k2.6","model_name":"Kimi K2.6","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.667,"normalized_score":0.667,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.75,"output_cost_per_million":3.5,"speed_rps":null,"context_window":262144,"release_date":"2026-04-20","announcement_date":"2026-04-20","multimodal":true,"param_count":1000000000000,"is_new":false},{"rank":16,"model_id":"mimo-v2.5","model_name":"MiMo-V2.5","organization_name":"Xiaomi","organization_id":"xiaomi","benchmark_score":0.658,"normalized_score":0.658,"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":17,"model_id":"claude-opus-4-6","model_name":"Claude Opus 4.6","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.654,"normalized_score":0.654,"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":18,"model_id":"gpt-5.2-codex","model_name":"GPT-5.2 Codex","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.64,"normalized_score":0.64,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":1.75,"output_cost_per_million":14.0,"speed_rps":50.0,"context_window":400000,"release_date":"2026-01-14","announcement_date":"2026-01-14","multimodal":true,"param_count":null,"is_new":false},{"rank":19,"model_id":"qwen3.6-plus","model_name":"Qwen3.6 Plus","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.616,"normalized_score":0.616,"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":20,"model_id":"gpt-5.4-mini","model_name":"GPT-5.4 mini","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.6,"normalized_score":0.6,"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":21,"model_id":"qwen3.6-27b","model_name":"Qwen3.6-27B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.593,"normalized_score":0.593,"verified":false,"self_reported":true,"provider_id":"novita","input_cost_per_million":0.6,"output_cost_per_million":3.6,"speed_rps":null,"context_window":262144,"release_date":"2026-04-21","announcement_date":"2026-04-21","multimodal":true,"param_count":27781427952,"is_new":false},{"rank":22,"model_id":"claude-opus-4-5-20251101","model_name":"Claude Opus 4.5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.593,"normalized_score":0.593,"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":23,"model_id":"claude-sonnet-4-6","model_name":"Claude Sonnet 4.6","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.591,"normalized_score":0.591,"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":24,"model_id":"muse-spark","model_name":"Muse Spark","organization_name":"Meta","organization_id":"meta","benchmark_score":0.59,"normalized_score":0.59,"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":25,"model_id":"mimo-v2-pro","model_name":"MiMo-V2-Pro","organization_name":"Xiaomi","organization_id":"xiaomi","benchmark_score":0.571,"normalized_score":0.571,"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":26,"model_id":"minimax-m2.7","model_name":"MiniMax M2.7","organization_name":"MiniMax","organization_id":"minimax","benchmark_score":0.57,"normalized_score":0.57,"verified":false,"self_reported":true,"provider_id":"fireworks","input_cost_per_million":0.3,"output_cost_per_million":1.2,"speed_rps":null,"context_window":196608,"release_date":"2026-03-18","announcement_date":"2026-03-18","multimodal":false,"param_count":null,"is_new":false},{"rank":27,"model_id":"deepseek-v4-flash-max","model_name":"DeepSeek-V4-Flash-Max","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.569,"normalized_score":0.569,"verified":false,"self_reported":true,"provider_id":"deepseek","input_cost_per_million":0.14,"output_cost_per_million":0.28,"speed_rps":null,"context_window":1048576,"release_date":"2026-04-23","announcement_date":"2026-04-23","multimodal":false,"param_count":284000000000,"is_new":false},{"rank":28,"model_id":"deepseek-v4-flash-0423","model_name":"DeepSeek-V4-Flash-0423","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.566,"normalized_score":0.566,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.1,"output_cost_per_million":0.2,"speed_rps":null,"context_window":1048576,"release_date":"2026-04-23","announcement_date":"2026-04-23","multimodal":false,"param_count":284000000000,"is_new":false},{"rank":29,"model_id":"glm-5","model_name":"GLM-5","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.562,"normalized_score":0.562,"verified":false,"self_reported":true,"provider_id":"friendli","input_cost_per_million":1.0,"output_cost_per_million":3.2,"speed_rps":null,"context_window":200000,"release_date":"2026-02-11","announcement_date":"2026-02-11","multimodal":false,"param_count":744000000000,"is_new":false},{"rank":30,"model_id":"mai-code-1-flash","model_name":"MAI-Code-1-Flash","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.548,"normalized_score":0.548,"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":31,"model_id":"gemini-3-pro-preview","model_name":"Gemini 3 Pro","organization_name":"Google","organization_id":"google","benchmark_score":0.542,"normalized_score":0.542,"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":"2025-11-18","announcement_date":"2025-11-18","multimodal":true,"param_count":null,"is_new":false},{"rank":32,"model_id":"gpt-5.1-codex","model_name":"GPT-5.1 Codex","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.528,"normalized_score":0.528,"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":"2025-11-19","announcement_date":"2025-11-19","multimodal":true,"param_count":null,"is_new":false},{"rank":33,"model_id":"qwen3.5-397b-a17b","model_name":"Qwen3.5-397B-A17B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.525,"normalized_score":0.525,"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":34,"model_id":"qwen3.6-35b-a3b","model_name":"Qwen3.6-35B-A3B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.515,"normalized_score":0.515,"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":"2026-04-16","announcement_date":"2026-04-16","multimodal":true,"param_count":35000000000,"is_new":false},{"rank":35,"model_id":"step-3.5-flash","model_name":"Step-3.5-Flash","organization_name":"StepFun","organization_id":"stepfun","benchmark_score":0.51,"normalized_score":0.51,"verified":false,"self_reported":true,"provider_id":"stepfun","input_cost_per_million":0.1,"output_cost_per_million":0.4,"speed_rps":150.0,"context_window":65536,"release_date":"2026-02-02","announcement_date":"2026-02-02","multimodal":false,"param_count":196000000000,"is_new":false},{"rank":36,"model_id":"kimi-k2.5","model_name":"Kimi K2.5","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.508,"normalized_score":0.508,"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":37,"model_id":"qwen3.5-122b-a10b","model_name":"Qwen3.5-122B-A10B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.494,"normalized_score":0.494,"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":38,"model_id":"gemini-3-flash-preview","model_name":"Gemini 3 Flash","organization_name":"Google","organization_id":"google","benchmark_score":0.476,"normalized_score":0.476,"verified":false,"self_reported":true,"provider_id":"google","input_cost_per_million":0.5,"output_cost_per_million":3.0,"speed_rps":null,"context_window":1000000,"release_date":"2025-12-17","announcement_date":"2025-12-17","multimodal":true,"param_count":null,"is_new":false},{"rank":39,"model_id":"deepseek-v3.2","model_name":"DeepSeek-V3.2","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.464,"normalized_score":0.464,"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":"2025-12-01","announcement_date":"2025-12-01","multimodal":false,"param_count":685000000000,"is_new":false},{"rank":40,"model_id":"deepseek-v3.2-speciale","model_name":"DeepSeek-V3.2-Speciale","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.464,"normalized_score":0.464,"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":"2025-12-01","announcement_date":"2025-12-01","multimodal":false,"param_count":685000000000,"is_new":false},{"rank":41,"model_id":"deepseek-reasoner","model_name":"DeepSeek-V3.2 (Thinking)","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.464,"normalized_score":0.464,"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":"2025-12-01","announcement_date":"2025-12-01","multimodal":false,"param_count":685000000000,"is_new":false},{"rank":42,"model_id":"gpt-5.4-nano","model_name":"GPT-5.4 nano","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.463,"normalized_score":0.463,"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":43,"model_id":"mai-thinking-1","model_name":"MAI-Thinking-1","organization_name":"Microsoft","organization_id":"microsoft","benchmark_score":0.46,"normalized_score":0.46,"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":44,"model_id":"qwen3.5-27b","model_name":"Qwen3.5-27B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.416,"normalized_score":0.416,"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":45,"model_id":"glm-4.7","model_name":"GLM-4.7","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.41,"normalized_score":0.41,"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":"2025-12-22","announcement_date":"2025-12-22","multimodal":true,"param_count":358000000000,"is_new":false},{"rank":46,"model_id":"qwen3.5-35b-a3b","model_name":"Qwen3.5-35B-A3B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.405,"normalized_score":0.405,"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":47,"model_id":"mimo-v2-flash","model_name":"MiMo-V2-Flash","organization_name":"Xiaomi","organization_id":"xiaomi","benchmark_score":0.385,"normalized_score":0.385,"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":"2025-12-16","announcement_date":"2025-12-16","multimodal":false,"param_count":309000000000,"is_new":false},{"rank":48,"model_id":"laguna-xs-2.1","model_name":"Laguna XS 2.1","organization_name":"Poolside","organization_id":"poolside","benchmark_score":0.375,"normalized_score":0.375,"verified":false,"self_reported":true,"provider_id":"poolside","input_cost_per_million":0.1,"output_cost_per_million":0.2,"speed_rps":null,"context_window":262144,"release_date":"2026-07-02","announcement_date":"2026-07-02","multimodal":false,"param_count":33000000000,"is_new":false},{"rank":49,"model_id":"qwen3-coder-480b-a35b-instruct","model_name":"Qwen3-Coder 480B A35B Instruct","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.375,"normalized_score":0.375,"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":"2025-01-31","announcement_date":"2025-01-31","multimodal":false,"param_count":480000000000,"is_new":false},{"rank":50,"model_id":"north-mini-code-1.0","model_name":"North Mini Code 1.0","organization_name":"Cohere","organization_id":"cohere","benchmark_score":0.36,"normalized_score":0.36,"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":"2026-06-09","announcement_date":"2026-06-09","multimodal":false,"param_count":30000000000,"is_new":false},{"rank":51,"model_id":"nemotron-3-super-120b-a12b","model_name":"Nemotron 3 Super (120B A12B)","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.31,"normalized_score":0.31,"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}]}