[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"aiModelCatalog":3},{"developers":4,"tags":49,"models":161},[5,9,13,17,21,25,29,33,37,41,45],{"code":6,"name":7,"logo_url":8},"anthropic","Anthropic","https://img.beget.com/b/plain/icons/anthropic-72-outlined.svg",{"code":10,"name":11,"logo_url":12},"deepseek","DeepSeek","https://img.beget.com/b/plain/icons/deepseek-72-outlined.svg",{"code":14,"name":15,"logo_url":16},"glm","GLM","https://img.beget.com/b/plain/icons/glm-72-outlined.svg",{"code":18,"name":19,"logo_url":20},"google","Google","https://img.beget.com/b/plain/icons/google-72-outlined.svg",{"code":22,"name":23,"logo_url":24},"kimi","Kimi","https://img.beget.com/b/plain/icons/kimi-72-outlined.svg",{"code":26,"name":27,"logo_url":28},"meta","Meta","https://img.beget.com/b/plain/icons/meta-72-outlined.svg",{"code":30,"name":31,"logo_url":32},"minimax","MiniMax","https://img.beget.com/b/plain/icons/minimax-72-outlined.svg",{"code":34,"name":35,"logo_url":36},"mistralai","Mistral AI","https://img.beget.com/b/plain/icons/mistralai-72-outlined.svg",{"code":38,"name":39,"logo_url":40},"openai","OpenAI","https://img.beget.com/b/plain/icons/openai-72-outlined.svg",{"code":42,"name":43,"logo_url":44},"qwen","Qwen","https://img.beget.com/b/plain/icons/qwen-72-outlined.svg",{"code":46,"name":47,"logo_url":48},"xai","xAI","https://img.beget.com/b/plain/icons/xai-72-outlined.svg",[50,65,76,87,98,110,121,133,144],{"code":51,"title":52,"tooltip":59,"tag_class":60,"is_filter":61,"filter_title":62},"universal-quality",[53,56],{"locale":54,"value":55},"EN","General-purpose",{"locale":57,"value":58},"RU","Универсальная",[],"CLASS_UNSPECIFIED",true,[63,64],{"locale":54,"value":55},{"locale":57,"value":58},{"code":66,"title":67,"tooltip":72,"tag_class":60,"is_filter":61,"filter_title":73},"cheap-fast",[68,70],{"locale":54,"value":69},"Fast and affordable",{"locale":57,"value":71},"Дешёвая быстрая",[],[74,75],{"locale":54,"value":69},{"locale":57,"value":71},{"code":77,"title":78,"tooltip":83,"tag_class":60,"is_filter":61,"filter_title":84},"long-context",[79,81],{"locale":54,"value":80},"Long context",{"locale":57,"value":82},"Длинный контекст",[],[85,86],{"locale":54,"value":80},{"locale":57,"value":82},{"code":88,"title":89,"tooltip":94,"tag_class":60,"is_filter":61,"filter_title":95},"complex-tasks",[90,92],{"locale":54,"value":91},"Complex tasks",{"locale":57,"value":93},"Сложные задачи",[],[96,97],{"locale":54,"value":91},{"locale":57,"value":93},{"code":99,"title":100,"tooltip":105,"tag_class":106,"is_filter":61,"filter_title":107},"experimental",[101,103],{"locale":54,"value":102},"Experimental",{"locale":57,"value":104},"Экспериментальная",[],"EXPERIMENTAL",[108,109],{"locale":54,"value":102},{"locale":57,"value":104},{"code":111,"title":112,"tooltip":117,"tag_class":60,"is_filter":61,"filter_title":118},"coding",[113,115],{"locale":54,"value":114},"Coding",{"locale":57,"value":116},"Кодинг",[],[119,120],{"locale":54,"value":114},{"locale":57,"value":116},{"code":122,"title":123,"tooltip":128,"tag_class":60,"is_filter":61,"filter_title":129},"price-quality",[124,126],{"locale":54,"value":125},"Value for money",{"locale":57,"value":127},"Баланс цена/качество",[],[130,131],{"locale":54,"value":125},{"locale":57,"value":132},"Цена/качество",{"code":134,"title":135,"tooltip":140,"tag_class":60,"is_filter":61,"filter_title":141},"agent-scenarios",[136,138],{"locale":54,"value":137},"Agent workflows",{"locale":57,"value":139},"Агентские сценарии",[],[142,143],{"locale":54,"value":137},{"locale":57,"value":139},{"code":145,"title":146,"tooltip":151,"tag_class":156,"is_filter":61,"filter_title":157},"beget-recommended",[147,149],{"locale":54,"value":148},"Recommended by Beget",{"locale":57,"value":150},"Рекомендовано Beget",[152,154],{"locale":54,"value":153},"Based on value for money and benchmark analysis",{"locale":57,"value":155},"По соотношению цена/качество и анализу бенчмарков","RECOMMENDED",[158,159],{"locale":54,"value":148},{"locale":57,"value":160},"Рекомендует Beget",[162,177,191,205,219,233,247,262,276,290,304,318,333,348,361,375,389,404,415,428,442,455,469,483,496,508,520,534,546,558],{"id":163,"display_name":164,"api_name":165,"type":166,"context_size":167,"input_price_per_million_tokens":168,"output_price_per_million_tokens":169,"active":61,"developer_code":38,"description":170,"tag":175},"1","GPT-5.6 Sol","gpt-5.6-sol","AI_MODEL_TYPE_CHAT",1050000,478,2388,[171,173],{"locale":54,"value":172},"OpenAI's previous-generation flagship for complex assistants, coding, and document analysis. The new model in this family is GPT-6.1 Sol.",{"locale":57,"value":174},"Предыдущее поколение флагмана OpenAI для сложных ассистентов, кода и анализа документов. Новая модель этой линейки — GPT-6.1 Sol.",[176],{"code":51,"is_visible":61},{"id":178,"display_name":179,"api_name":180,"type":166,"context_size":167,"input_price_per_million_tokens":181,"output_price_per_million_tokens":182,"active":61,"developer_code":38,"description":183,"tag":188},"2","GPT-5.6 Luna","gpt-5.6-luna",24,144,[184,186],{"locale":54,"value":185},"OpenAI's lightweight model family for classification, short responses, and batch requests. The new model in this family is GPT-6 Luna.",{"locale":57,"value":187},"Лёгкая линейка OpenAI для классификации, коротких ответов и массовых запросов. Новая модель этой линейки — GPT-6 Luna.",[189,190],{"code":66,"is_visible":61},{"code":122,"is_visible":61},{"id":192,"display_name":193,"api_name":194,"type":166,"context_size":195,"input_price_per_million_tokens":196,"output_price_per_million_tokens":197,"active":61,"developer_code":6,"description":198,"tag":203},"3","Claude Sonnet 5","claude-sonnet-5",1000000,239,1194,[199,201],{"locale":54,"value":200},"Anthropic's workhorse for instructions, coding, analytics, and complex conversations. The new model in this family is Sonnet 5.5.",{"locale":57,"value":202},"Рабочая модель Anthropic для инструкций, кода, аналитики и сложных диалогов. Новая модель этой линейки — Sonnet 5.5.",[204],{"code":51,"is_visible":61},{"id":206,"display_name":207,"api_name":208,"type":166,"context_size":195,"input_price_per_million_tokens":209,"output_price_per_million_tokens":210,"active":61,"developer_code":6,"description":211,"tag":216},"4","Claude Opus 5","claude-opus-5",597,2984,[212,214],{"locale":54,"value":213},"Anthropic's model for multi-step reasoning, architectural decisions, and complex agentic workflows. The new model in this family is Opus 5.5.",{"locale":57,"value":215},"Модель Anthropic для многошаговых рассуждений, архитектурных решений и сложных агентских процессов. Новая модель этой линейки — Opus 5.5.",[217,218],{"code":88,"is_visible":61},{"code":51,"is_visible":61},{"id":220,"display_name":221,"api_name":222,"type":166,"context_size":223,"input_price_per_million_tokens":196,"output_price_per_million_tokens":224,"active":61,"developer_code":18,"description":225,"tag":230},"5","Gemini 3.1 Pro Preview","gemini-3.1-pro",1048576,1433,[226,228],{"locale":54,"value":227},"Google's Pro model: analyzing long documents, knowledge bases, and conversations with a large 1M context.",{"locale":57,"value":229},"Pro-линейка Google: анализ длинных документов, knowledge bases и диалоги с большим контекстом 1M.",[231,232],{"code":77,"is_visible":61},{"code":145,"is_visible":61},{"id":234,"display_name":235,"api_name":236,"type":166,"context_size":223,"input_price_per_million_tokens":237,"output_price_per_million_tokens":238,"active":61,"developer_code":18,"description":239,"tag":244},"6","Gemini 3.1 Flash Lite Preview","gemini-3.1-flash-lite",30,180,[240,242],{"locale":54,"value":241},"Flash Lite: low latency and cost with a 1M context — for classification and high-throughput workloads.",{"locale":57,"value":243},"Flash Lite: низкая latency и стоимость при контексте 1M — classification и массовый high-throughput.",[245,246],{"code":66,"is_visible":61},{"code":77,"is_visible":61},{"id":248,"display_name":249,"api_name":250,"type":166,"context_size":251,"input_price_per_million_tokens":252,"output_price_per_million_tokens":253,"active":61,"developer_code":10,"description":254,"tag":259},"7","DeepSeek V4 Pro","deepseek-v4-pro",600000,138,305,[255,257],{"locale":54,"value":256},"Strong quality at a low cost — a go-to choice for results without a premium budget.",{"locale":57,"value":258},"Сильное качество при низкой стоимости — default-выбор, когда нужен результат без премиум-бюджета.",[260,261],{"code":122,"is_visible":61},{"code":145,"is_visible":61},{"id":263,"display_name":264,"api_name":265,"type":166,"context_size":223,"input_price_per_million_tokens":266,"output_price_per_million_tokens":267,"active":61,"developer_code":10,"description":268,"tag":273},"8","DeepSeek V4 Flash","deepseek-v4-flash",4,7,[269,271],{"locale":54,"value":270},"A high-throughput model for simple responses, classification, and large request volumes.",{"locale":57,"value":272},"High-throughput модель для простых ответов, классификации и большого объёма запросов.",[274,275],{"code":66,"is_visible":61},{"code":145,"is_visible":61},{"id":277,"display_name":278,"api_name":279,"type":166,"context_size":280,"input_price_per_million_tokens":281,"output_price_per_million_tokens":282,"active":61,"developer_code":10,"description":283,"tag":288},"9","DeepSeek R1 0528","deepseek-r1",131072,84,299,[284,286],{"locale":54,"value":285},"A reasoning model for planning, multi-step analysis, and tasks that require a chain of reasoning.",{"locale":57,"value":287},"Reasoning-модель: планирование, multi-step анализ и задачи, где важна цепочка рассуждений.",[289],{"code":88,"is_visible":61},{"id":291,"display_name":292,"api_name":293,"type":166,"context_size":195,"input_price_per_million_tokens":294,"output_price_per_million_tokens":295,"active":61,"developer_code":42,"description":296,"tag":301},"10","Qwen3.7 Max","qwen-3.7-max",177,529,[297,299],{"locale":54,"value":298},"A general-purpose Max model with a 1M context and strong Russian-language support.",{"locale":57,"value":300},"Универсальная Max-модель с контекстом 1M и сильной поддержкой русского языка.",[302,303],{"code":51,"is_visible":61},{"code":77,"is_visible":61},{"id":305,"display_name":306,"api_name":307,"type":166,"context_size":308,"input_price_per_million_tokens":309,"output_price_per_million_tokens":310,"active":61,"developer_code":42,"description":311,"tag":316},"11","Qwen3 Coder Next","qwen3-coder-next",256000,15,96,[312,314],{"locale":54,"value":313},"A specialized coding model for code generation, review, explanation, and automated tests.",{"locale":57,"value":315},"Специализированная coding-модель: генерация, ревью, объяснение кода и автотесты.",[317],{"code":111,"is_visible":61},{"id":319,"display_name":320,"api_name":321,"type":166,"context_size":322,"input_price_per_million_tokens":323,"output_price_per_million_tokens":324,"active":61,"developer_code":22,"description":325,"tag":330},"12","Kimi K2.6","kimi-k2.6",262144,52,382,[326,328],{"locale":54,"value":327},"A model for agent and research workflows: multi-step reasoning, tool calling, and a 262K context.",{"locale":57,"value":329},"Модель для агентских и research-сценариев: multi-step reasoning, tool-calling, контекст 262K.",[331,332],{"code":77,"is_visible":61},{"code":134,"is_visible":61},{"id":334,"display_name":335,"api_name":336,"type":166,"context_size":337,"input_price_per_million_tokens":338,"output_price_per_million_tokens":339,"active":61,"developer_code":14,"description":340,"tag":345},"13","GLM 5.2","glm-5.2",202752,23,955,[341,343],{"locale":54,"value":342},"An open-weight model for agent pipelines and multi-step tasks. Good value for money.",{"locale":57,"value":344},"Open-weight модель для агентских пайплайнов и multi-step задач. Хороший баланс цены и качества.",[346,347],{"code":122,"is_visible":61},{"code":134,"is_visible":61},{"id":349,"display_name":350,"api_name":351,"type":166,"context_size":195,"input_price_per_million_tokens":352,"output_price_per_million_tokens":182,"active":61,"developer_code":30,"description":353,"tag":358},"14","MiniMax M3","minimax-m3",36,[354,356],{"locale":54,"value":355},"An affordable model with a 1M context for high-throughput tasks and long inputs.",{"locale":57,"value":357},"Бюджетная модель с контекстом 1M для high-throughput задач и длинных входных данных.",[359,360],{"code":66,"is_visible":61},{"code":77,"is_visible":61},{"id":362,"display_name":363,"api_name":364,"type":166,"context_size":223,"input_price_per_million_tokens":365,"output_price_per_million_tokens":366,"active":61,"developer_code":26,"description":367,"tag":372},"15","Llama 4 Maverick","llama-4-maverick",42,120,[368,370],{"locale":54,"value":369},"Meta's open-weight model — a familiar stack for experiments and general-purpose tasks.",{"locale":57,"value":371},"Open-weight модель Meta — знакомый стек для экспериментов и универсальных задач.",[373,374],{"code":51,"is_visible":61},{"code":99,"is_visible":61},{"id":376,"display_name":377,"api_name":378,"type":166,"context_size":379,"input_price_per_million_tokens":380,"output_price_per_million_tokens":282,"active":61,"developer_code":46,"description":381,"tag":386},"17","Grok 4.20","grok-4.20",2000000,150,[382,384],{"locale":54,"value":383},"A record 2M context — for research and analyzing entire logs and long conversations.",{"locale":57,"value":385},"Рекордный контекст 2M — research, анализ логов и длинных переписок целиком.",[387,388],{"code":99,"is_visible":61},{"code":77,"is_visible":61},{"id":390,"display_name":391,"api_name":391,"type":392,"context_size":393,"input_price_per_million_tokens":394,"output_price_per_million_tokens":395,"active":61,"developer_code":38,"description":396,"tag":401},"18","text-embedding-3-small","AI_MODEL_TYPE_EMBEDDING",8192,3,0,[397,399],{"locale":54,"value":398},"Compact OpenAI embeddings: semantic search, RAG, and clustering at a low cost.",{"locale":57,"value":400},"Компактные эмбеддинги OpenAI: семантический поиск, RAG и кластеризация при низкой стоимости.",[402,403],{"code":66,"is_visible":61},{"code":122,"is_visible":61},{"id":405,"display_name":406,"api_name":406,"type":392,"context_size":393,"input_price_per_million_tokens":407,"output_price_per_million_tokens":395,"active":61,"developer_code":38,"description":408,"tag":413},"19","text-embedding-3-large",16,[409,411],{"locale":54,"value":410},"Higher-dimensional OpenAI embeddings — maximum accuracy for semantic search and RAG over large text corpora.",{"locale":57,"value":412},"Эмбеддинги OpenAI повышенной размерности — максимальная точность семантического поиска и RAG на больших корпусах.",[414],{"code":51,"is_visible":61},{"id":416,"display_name":417,"api_name":418,"type":166,"context_size":195,"input_price_per_million_tokens":197,"output_price_per_million_tokens":419,"active":61,"developer_code":6,"description":420,"tag":425},"20","Claude Fable 5.1","claude-fable-5.1",5968,[421,423],{"locale":54,"value":422},"Anthropic's flagship: the highest Intelligence and Coding Index scores for long agent and reasoning workflows.",{"locale":57,"value":424},"Флагман Anthropic: максимальный Intelligence/Coding Index для длинных агентских и reasoning-сценариев.",[426,427],{"code":88,"is_visible":61},{"code":51,"is_visible":61},{"id":429,"display_name":430,"api_name":431,"type":166,"context_size":223,"input_price_per_million_tokens":432,"output_price_per_million_tokens":433,"active":61,"developer_code":18,"description":434,"tag":439},"21","Gemini 3.8 Flash","gemini-3.8-flash",90,448,[435,437],{"locale":54,"value":436},"Google's Flash model: high quality at a low price with a 1M context — fast responses without a major drop in quality.",{"locale":57,"value":438},"Flash Google: высокое качество при низкой цене, контекст 1M — быстрые ответы без сильной деградации.",[440,441],{"code":66,"is_visible":61},{"code":77,"is_visible":61},{"id":443,"display_name":444,"api_name":445,"type":166,"context_size":195,"input_price_per_million_tokens":196,"output_price_per_million_tokens":446,"active":61,"developer_code":42,"description":447,"tag":452},"22","Qwen3.8 Max","qwen3.8-max",717,[448,450],{"locale":54,"value":449},"Qwen's current Max model: general-purpose tasks, Russian-language support, and a long 1M context.",{"locale":57,"value":451},"Актуальный Max Qwen: универсальные задачи, русский язык, длинный контекст 1M.",[453,454],{"code":51,"is_visible":61},{"code":77,"is_visible":61},{"id":456,"display_name":457,"api_name":458,"type":166,"context_size":223,"input_price_per_million_tokens":459,"output_price_per_million_tokens":460,"active":61,"developer_code":22,"description":461,"tag":466},"23","Kimi K3","kimi-k3",359,1791,[462,464],{"locale":54,"value":463},"Kimi for agent pipelines and tool calling with a long 1M context.",{"locale":57,"value":465},"Kimi для агентских пайплайнов и tool-calling на длинном контексте 1M.",[467,468],{"code":134,"is_visible":61},{"code":77,"is_visible":61},{"id":470,"display_name":471,"api_name":472,"type":166,"context_size":223,"input_price_per_million_tokens":473,"output_price_per_million_tokens":474,"active":61,"developer_code":14,"description":475,"tag":480},"24","GLM 5.3","glm-5.3",168,526,[476,478],{"locale":54,"value":477},"GLM for agent workflows and tool calling. A 1M context and a strong Coding Index at a moderate price.",{"locale":57,"value":479},"GLM для агентских сценариев и tool-calling. Контекст 1M, сильный Coding Index при умеренной цене.",[481,482],{"code":134,"is_visible":61},{"code":122,"is_visible":61},{"id":484,"display_name":485,"api_name":486,"type":166,"context_size":487,"input_price_per_million_tokens":196,"output_price_per_million_tokens":446,"active":61,"developer_code":46,"description":488,"tag":493},"25","Grok 4.6","grok-4.6",500000,[489,491],{"locale":54,"value":490},"xAI's general-purpose model for reasoning and coding with a 500,000-token context window. The new model in this family is Grok 4.7.",{"locale":57,"value":492},"Универсальная модель xAI для рассуждений и кода с контекстом 500 тысяч токенов. Новая модель этой линейки — Grok 4.7.",[494,495],{"code":51,"is_visible":61},{"code":99,"is_visible":61},{"id":497,"display_name":498,"api_name":499,"type":166,"context_size":167,"input_price_per_million_tokens":196,"output_price_per_million_tokens":197,"active":61,"developer_code":38,"description":500,"tag":505},"26","GPT-6.1 Sol","gpt-6.1-sol",[501,503],{"locale":54,"value":502},"OpenAI's next-generation flagship for agentic coding, document analysis, and multi-step workflows.",{"locale":57,"value":504},"Флагман OpenAI нового поколения для агентного кода, анализа документов и многошаговых процессов.",[506,507],{"code":51,"is_visible":61},{"code":145,"is_visible":61},{"id":509,"display_name":510,"api_name":511,"type":166,"context_size":167,"input_price_per_million_tokens":197,"output_price_per_million_tokens":419,"active":61,"developer_code":38,"description":512,"tag":517},"27","GPT-6 Astra","gpt-6-astra",[513,515],{"locale":54,"value":514},"OpenAI's model for complex analysis, engineering, research, and long-running agentic tasks.",{"locale":57,"value":516},"Модель OpenAI для сложного анализа, инженерии и исследований с длинными агентными задачами.",[518,519],{"code":88,"is_visible":61},{"code":51,"is_visible":61},{"id":521,"display_name":522,"api_name":523,"type":166,"context_size":167,"input_price_per_million_tokens":524,"output_price_per_million_tokens":525,"active":61,"developer_code":38,"description":526,"tag":531},"28","GPT-6 Luna","gpt-6-luna",12,60,[527,529],{"locale":54,"value":528},"OpenAI's lightweight next-generation model for conversations, classification, and low-cost batch requests.",{"locale":57,"value":530},"Лёгкая модель OpenAI нового поколения для диалогов, классификации и массовых запросов с низкой стоимостью.",[532,533],{"code":66,"is_visible":61},{"code":122,"is_visible":61},{"id":535,"display_name":536,"api_name":537,"type":166,"context_size":195,"input_price_per_million_tokens":196,"output_price_per_million_tokens":197,"active":61,"developer_code":6,"description":538,"tag":543},"29","Claude Sonnet 5.5","claude-sonnet-5.5",[539,541],{"locale":54,"value":540},"Anthropic's next-generation workhorse for coding, instructions, business documents, and analytics.",{"locale":57,"value":542},"Рабочая модель Anthropic нового поколения для кода, инструкций, деловых документов и аналитики.",[544,545],{"code":51,"is_visible":61},{"code":145,"is_visible":61},{"id":547,"display_name":548,"api_name":549,"type":166,"context_size":195,"input_price_per_million_tokens":168,"output_price_per_million_tokens":169,"active":61,"developer_code":6,"description":550,"tag":555},"30","Claude Opus 5.5","claude-opus-5.5",[551,553],{"locale":54,"value":552},"Anthropic's model for complex tasks, in-depth analysis, and long-running agentic workflows.",{"locale":57,"value":554},"Модель Anthropic для сложных задач, глубокого анализа и длинных агентных процессов.",[556,557],{"code":88,"is_visible":61},{"code":51,"is_visible":61},{"id":559,"display_name":560,"api_name":561,"type":166,"context_size":487,"input_price_per_million_tokens":196,"output_price_per_million_tokens":446,"active":61,"developer_code":46,"description":562,"tag":567},"31","Grok 4.7","grok-4.7",[563,565],{"locale":54,"value":564},"xAI's next-generation general-purpose model for reasoning and coding with a 500,000-token context window.",{"locale":57,"value":566},"Универсальная модель xAI нового поколения для рассуждений и кода с контекстом 500 тысяч токенов.",[568,569],{"code":51,"is_visible":61},{"code":99,"is_visible":61}]