{"benchmark_id":"browsecomp","benchmark_name":"BrowseComp","benchmark_description":"BrowseComp is a benchmark comprising 1,266 questions that challenge AI agents to persistently navigate the internet in search of hard-to-find, entangled information. The benchmark measures agents' ability to exercise persistence in information gathering, demonstrate creativity in web navigation, and find concise, verifiable answers. Despite the difficulty of the questions, BrowseComp is simple and easy-to-use, as predicted answers are short and easily verifiable against reference answers.","max_score":1.0,"categories":["reasoning","search","agents"],"modality":"text","total_models":62,"entries":[{"rank":1,"model_id":"kimi-k3","model_name":"Kimi K3","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.912,"normalized_score":0.912,"verified":false,"self_reported":true,"provider_id":"novita","input_cost_per_million":3.0,"output_cost_per_million":15.0,"speed_rps":null,"context_window":1048576,"release_date":"2026-07-16","announcement_date":"2026-07-16","multimodal":true,"param_count":2800000000000,"is_new":false},{"rank":2,"model_id":"claude-opus-5","model_name":"Claude Opus 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.908,"normalized_score":0.908,"verified":false,"self_reported":true,"provider_id":"anthropic","input_cost_per_million":5.0,"output_cost_per_million":25.0,"speed_rps":null,"context_window":1000000,"release_date":"2026-07-24","announcement_date":"2026-07-24","multimodal":true,"param_count":null,"is_new":false},{"rank":3,"model_id":"gpt-5.6-sol","model_name":"GPT-5.6 Sol","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.904,"normalized_score":0.904,"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":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":4,"model_id":"gpt-5.5-pro","model_name":"GPT-5.5 Pro","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.901,"normalized_score":0.901,"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":true,"param_count":null,"is_new":false},{"rank":5,"model_id":"gpt-5.6-terra","model_name":"GPT-5.6 Terra","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.875,"normalized_score":0.875,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":2.0,"output_cost_per_million":12.0,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":6,"model_id":"claude-mythos-preview","model_name":"Claude Mythos Preview","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.869,"normalized_score":0.869,"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":7,"model_id":"kimi-k2.6","model_name":"Kimi K2.6","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.863,"normalized_score":0.863,"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":8,"model_id":"seed-2.1-pro","model_name":"Seed 2.1 Pro","organization_name":"ByteDance","organization_id":"bytedance","benchmark_score":0.862,"normalized_score":0.862,"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":"2026-06-24","announcement_date":"2026-06-24","multimodal":true,"param_count":null,"is_new":false},{"rank":9,"model_id":"gemini-3.1-pro-preview","model_name":"Gemini 3.1 Pro","organization_name":"Google","organization_id":"google","benchmark_score":0.859,"normalized_score":0.859,"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":10,"model_id":"seed-2.1-turbo","model_name":"Seed 2.1 Turbo","organization_name":"ByteDance","organization_id":"bytedance","benchmark_score":0.849,"normalized_score":0.849,"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":"2026-06-24","announcement_date":"2026-06-24","multimodal":true,"param_count":null,"is_new":false},{"rank":11,"model_id":"claude-sonnet-5","model_name":"Claude Sonnet 5","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.847,"normalized_score":0.847,"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":12,"model_id":"gpt-5.5","model_name":"GPT-5.5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.844,"normalized_score":0.844,"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":13,"model_id":"claude-opus-4-8","model_name":"Claude Opus 4.8","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.843,"normalized_score":0.843,"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":14,"model_id":"hy3","model_name":"Hy3","organization_name":"Tencent","organization_id":"tencent","benchmark_score":0.842,"normalized_score":0.842,"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":15,"model_id":"claude-opus-4-6","model_name":"Claude Opus 4.6","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.84,"normalized_score":0.84,"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":16,"model_id":"minimax-m3","model_name":"MiniMax M3","organization_name":"MiniMax","organization_id":"minimax","benchmark_score":0.8352,"normalized_score":0.8352,"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":1000000,"release_date":"2026-06-01","announcement_date":"2026-06-01","multimodal":true,"param_count":428000000000,"is_new":false},{"rank":17,"model_id":"deepseek-v4-pro-max","model_name":"DeepSeek-V4-Pro-Max","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.834,"normalized_score":0.834,"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":18,"model_id":"gpt-5.6-luna","model_name":"GPT-5.6 Luna","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.833,"normalized_score":0.833,"verified":false,"self_reported":true,"provider_id":"openai","input_cost_per_million":0.2,"output_cost_per_million":1.2,"speed_rps":null,"context_window":1050000,"release_date":"2026-07-09","announcement_date":"2026-07-09","multimodal":true,"param_count":null,"is_new":false},{"rank":19,"model_id":"gpt-5.4","model_name":"GPT-5.4","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":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":20,"model_id":"claude-opus-4-7","model_name":"Claude Opus 4.7","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.793,"normalized_score":0.793,"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":21,"model_id":"glm-5.1","model_name":"GLM-5.1","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.793,"normalized_score":0.793,"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":22,"model_id":"gpt-5.2-pro-2025-12-11","model_name":"GPT-5.2 Pro","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.779,"normalized_score":0.779,"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":"2025-12-11","announcement_date":"2025-12-11","multimodal":true,"param_count":null,"is_new":false},{"rank":23,"model_id":"inkling-small","model_name":"Inkling-Small","organization_name":"Thinking Machines Lab","organization_id":"thinking-machines","benchmark_score":0.774,"normalized_score":0.774,"verified":false,"self_reported":true,"provider_id":"thinking-machines","input_cost_per_million":0.3,"output_cost_per_million":1.2,"speed_rps":null,"context_window":256000,"release_date":"2026-07-30","announcement_date":"2026-07-30","multimodal":true,"param_count":276000000000,"is_new":false},{"rank":24,"model_id":"seed-2.0-pro","model_name":"Seed 2.0 Pro","organization_name":"ByteDance","organization_id":"bytedance","benchmark_score":0.773,"normalized_score":0.773,"verified":false,"self_reported":true,"provider_id":"deepinfra","input_cost_per_million":0.5,"output_cost_per_million":3.0,"speed_rps":null,"context_window":256000,"release_date":"2026-02-14","announcement_date":"2026-02-14","multimodal":true,"param_count":null,"is_new":false},{"rank":25,"model_id":"minimax-m2.5","model_name":"MiniMax M2.5","organization_name":"MiniMax","organization_id":"minimax","benchmark_score":0.763,"normalized_score":0.763,"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":"2026-02-12","announcement_date":"2026-02-12","multimodal":false,"param_count":230000000000,"is_new":false},{"rank":26,"model_id":"glm-5","model_name":"GLM-5","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.759,"normalized_score":0.759,"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":27,"model_id":"kimi-k2.5","model_name":"Kimi K2.5","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.749,"normalized_score":0.749,"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":28,"model_id":"claude-sonnet-4-6","model_name":"Claude Sonnet 4.6","organization_name":"Anthropic","organization_id":"anthropic","benchmark_score":0.747,"normalized_score":0.747,"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":29,"model_id":"deepseek-v4-flash-max","model_name":"DeepSeek-V4-Flash-Max","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.732,"normalized_score":0.732,"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":30,"model_id":"qwen3.5-397b-a17b","model_name":"Qwen3.5-397B-A17B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.69,"normalized_score":0.69,"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":31,"model_id":"step-3.5-flash","model_name":"Step-3.5-Flash","organization_name":"StepFun","organization_id":"stepfun","benchmark_score":0.69,"normalized_score":0.69,"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":32,"model_id":"gpt-5.2-2025-12-11","model_name":"GPT-5.2","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.658,"normalized_score":0.658,"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":33,"model_id":"qwen3.5-122b-a10b","model_name":"Qwen3.5-122B-A10B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.638,"normalized_score":0.638,"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":34,"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":35,"model_id":"qwen3.5-27b","model_name":"Qwen3.5-27B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.61,"normalized_score":0.61,"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":36,"model_id":"qwen3.5-35b-a3b","model_name":"Qwen3.5-35B-A3B","organization_name":"Alibaba Cloud / Qwen Team","organization_id":"qwen","benchmark_score":0.61,"normalized_score":0.61,"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":37,"model_id":"kimi-k2-thinking-0905","model_name":"Kimi K2-Thinking-0905","organization_name":"Moonshot AI","organization_id":"moonshotai","benchmark_score":0.602,"normalized_score":0.602,"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":38,"model_id":"mimo-v2-flash","model_name":"MiMo-V2-Flash","organization_name":"Xiaomi","organization_id":"xiaomi","benchmark_score":0.583,"normalized_score":0.583,"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":39,"model_id":"longcat-flash-thinking-2601","model_name":"LongCat-Flash-Thinking-2601","organization_name":"Meituan","organization_id":"meituan","benchmark_score":0.566,"normalized_score":0.566,"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":40,"model_id":"gpt-5-2025-08-07","model_name":"GPT-5","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.549,"normalized_score":0.549,"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":41,"model_id":"deepseek-v4-flash-0423","model_name":"DeepSeek-V4-Flash-0423","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.535,"normalized_score":0.535,"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":42,"model_id":"glm-4.7","model_name":"GLM-4.7","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.52,"normalized_score":0.52,"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":43,"model_id":"o4-mini","model_name":"o4-mini","organization_name":"OpenAI","organization_id":"openai","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":"2025-04-16","announcement_date":"2025-04-16","multimodal":true,"param_count":null,"is_new":false},{"rank":44,"model_id":"deepseek-reasoner","model_name":"DeepSeek-V3.2 (Thinking)","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.514,"normalized_score":0.514,"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":45,"model_id":"deepseek-v3.2","model_name":"DeepSeek-V3.2","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.514,"normalized_score":0.514,"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":46,"model_id":"o3-2025-04-16","model_name":"o3","organization_name":"OpenAI","organization_id":"openai","benchmark_score":0.497,"normalized_score":0.497,"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":47,"model_id":"sarvam-105b","model_name":"Sarvam-105B","organization_name":"Sarvam AI","organization_id":"sarvamai","benchmark_score":0.495,"normalized_score":0.495,"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":"2026-03-06","announcement_date":"2026-03-06","multimodal":false,"param_count":105000000000,"is_new":false},{"rank":48,"model_id":"solar-pro4","model_name":"Solar Pro 4","organization_name":"Upstage","organization_id":"upstage","benchmark_score":0.492,"normalized_score":0.492,"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":49,"model_id":"mistral-medium-3-5","model_name":"Mistral Medium 3.5","organization_name":"Mistral AI","organization_id":"mistral","benchmark_score":0.486,"normalized_score":0.486,"verified":false,"self_reported":true,"provider_id":"mistral","input_cost_per_million":1.5,"output_cost_per_million":7.5,"speed_rps":null,"context_window":256000,"release_date":"2026-04-29","announcement_date":"2026-04-29","multimodal":true,"param_count":128000000000,"is_new":false},{"rank":50,"model_id":"glm-4.6","model_name":"GLM-4.6","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.451,"normalized_score":0.451,"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":"2025-09-30","announcement_date":"2025-09-30","multimodal":true,"param_count":357000000000,"is_new":false},{"rank":51,"model_id":"grok-4-fast","model_name":"Grok 4 Fast","organization_name":"xAI","organization_id":"xai","benchmark_score":0.449,"normalized_score":0.449,"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":"2025-08-28","announcement_date":"2025-08-28","multimodal":true,"param_count":null,"is_new":false},{"rank":52,"model_id":"nemotron-3-ultra-550b-a55b","model_name":"Nemotron 3 Ultra (550B A55B)","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.444,"normalized_score":0.444,"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":53,"model_id":"minimax-m2","model_name":"MiniMax M2","organization_name":"MiniMax","organization_id":"minimax","benchmark_score":0.44,"normalized_score":0.44,"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":54,"model_id":"glm-4.7-flash","model_name":"GLM-4.7-Flash","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.428,"normalized_score":0.428,"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":"2026-01-19","announcement_date":"2026-01-19","multimodal":false,"param_count":30000000000,"is_new":false},{"rank":55,"model_id":"deepseek-v3.2-exp","model_name":"DeepSeek-V3.2-Exp","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.401,"normalized_score":0.401,"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":"2025-09-29","announcement_date":"2025-09-29","multimodal":false,"param_count":685000000000,"is_new":false},{"rank":56,"model_id":"nemotron-3.5-lightning-30b-a3b","model_name":"Nemotron 3.5 Lightning (30B A3B)","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.3697,"normalized_score":0.3697,"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":57,"model_id":"sarvam-30b","model_name":"Sarvam-30B","organization_name":"Sarvam AI","organization_id":"sarvamai","benchmark_score":0.355,"normalized_score":0.355,"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":"2026-03-06","announcement_date":"2026-03-06","multimodal":false,"param_count":30000000000,"is_new":false},{"rank":58,"model_id":"nemotron-3-super-120b-a12b","model_name":"Nemotron 3 Super (120B A12B)","organization_name":"NVIDIA","organization_id":"nvidia","benchmark_score":0.3128,"normalized_score":0.3128,"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":59,"model_id":"deepseek-v3.1","model_name":"DeepSeek-V3.1","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.3,"normalized_score":0.3,"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":"2025-01-10","announcement_date":"2025-01-10","multimodal":false,"param_count":671000000000,"is_new":false},{"rank":60,"model_id":"glm-4.5","model_name":"GLM-4.5","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.264,"normalized_score":0.264,"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":"2025-07-28","announcement_date":"2025-07-28","multimodal":false,"param_count":355000000000,"is_new":false},{"rank":61,"model_id":"glm-4.5-air","model_name":"GLM-4.5-Air","organization_name":"Zhipu AI","organization_id":"zai-org","benchmark_score":0.213,"normalized_score":0.213,"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":"2025-07-28","announcement_date":"2025-07-28","multimodal":false,"param_count":106000000000,"is_new":false},{"rank":62,"model_id":"deepseek-r1-0528","model_name":"DeepSeek-R1-0528","organization_name":"DeepSeek","organization_id":"deepseek","benchmark_score":0.089,"normalized_score":0.089,"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":"2025-05-28","announcement_date":"2025-05-28","multimodal":false,"param_count":671000000000,"is_new":false}]}