{"auto_groups":["default"],"data":[{"context_length_display":"500K","max_output_tokens_display":"500K","model_name":"grok-4.7","created_time":1790057842,"description":"Grok 4.7 is SpaceXAI's flagship model for coding, agentic tasks, and knowledge work, succeeding Grok 4.6. It is particularly strong at long-running software engineering tasks, verifying its own work, and managing long context, and it improves on its predecessor at professional knowledge work such as drafting documents and presentations.\n\nThe model was trained with a longer reinforcement learning run weighted toward problems that take many hours to complete, and natively understands the Grok Bot harness for conversational tasks. It ships with a new safeguard stack that pairs strong jailbreak resistance with low refusal rates for legitimate cybersecurity and biology work. SpaceXAI's reported benchmark results use the xhigh reasoning effort.","tags":"Chat","vendor_id":14,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image"],"output_modalities":["text"],"vendor_name":"xAI","vendor_icon":"https://t0.gstatic.com/faviconV2?client=SOCIAL\u0026type=FAVICON\u0026fallback_opts=TYPE,SIZE,URL\u0026url=https://x.ai/\u0026size=256","billing_mode":"tiered_expr","billing_expr":"len \u003c= 200000 ? tier(\"standard\", p * 2 + c * 6 + cr * 0.5) : tier(\"long_context\", p * 4 + c * 12 + cr * 1)","pricing_version":"5a90f2b86c08bd983a9a2e6d66c255f4eaef9c4bc934386d2b6ae84ef0ff1f1f","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"1M","max_output_tokens_display":"128K","usage_tokens":34,"model_name":"mimo-v2.6-flash","created_time":1790035188,"description":"MiMo-V2.6-Flash is an open-source foundation model developed by Xiaomi. Built on a Mixture-of-Experts architecture with 309B total parameters and 15B activated per token, it employs a hybrid attention mechanism for greater computational efficiency. The model features a 1M-token context window and native multimodal capabilities. Optimized for agentic workflows, it delivers strong performance across coding, visual, general, and research scenarios, excelling at complex, long-horizon tasks with robust generalization across a diverse range of agent harnesses.","tags":"Chat","vendor_id":41,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","audio","video"],"output_modalities":["text"],"vendor_name":"Xiaomi","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/80caddd717cfafd45b0671c5b8969296_1779343536965.svg","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 0.14 + c * 0.28 + cr * 0.0028)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"1M","max_output_tokens_display":"128K","usage_tokens":34,"model_name":"mimo-v2.6-pro","created_time":1790035145,"description":"MiMo-V2.6-Pro is the flagship foundation model developed by Xiaomi. Built at a scale of over 1T parameters, it is designed to push the ceiling of capability for the most demanding workloads. The model features a 1M-token context window and native multimodal capabilities. Optimized for agentic workflows, it delivers top-tier performance across coding, visual, general, and research scenarios, excelling at complex, long-horizon tasks with robust generalization across a diverse range of agent harnesses.","tags":"Chat","vendor_id":41,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","audio","video"],"output_modalities":["text"],"vendor_name":"Xiaomi","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/80caddd717cfafd45b0671c5b8969296_1779343536965.svg","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 0.435 + c * 0.87 + cr * 0.0036)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"1M","max_output_tokens_display":"128K","usage_tokens":34,"model_name":"mimo-v2.6-pro-ultraspeed","created_time":1790035103,"description":"MiMo-V2.6-Pro-UltraSpeed is the fast speed edition of Xiaomi's flagship foundation model, MiMo-V2.6-Pro. Built from the same 1T MiMo-V2.6-Pro checkpoint, it matches the original model in quality while delivering roughly 10x the output speed. The model features a 1M-token context window and native multimodal capabilities. Optimized for agentic workflows, it delivers top-tier performance across coding, visual, general, and research scenarios, excelling at complex, long-horizon tasks with robust generalization across a diverse range of agent harnesses.","tags":"Chat","vendor_id":41,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","audio","video"],"output_modalities":["text"],"vendor_name":"Xiaomi","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/80caddd717cfafd45b0671c5b8969296_1779343536965.svg","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 4.35 + c * 8.7 + cr * 0.036)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"laya","created_time":1790015706,"description":"laya is an open-source, multilingual, non-autoregressive System 1 decision engine, designed for efficient and precise structured decision-making tasks. Through a single forward pass, it directly outputs typed decision results without generating natural-language text, thereby avoiding format hallucinations and parsing errors.","vendor_id":47,"quota_type":0,"model_ratio":0,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai","systemone"],"input_modalities":["text"],"output_modalities":["decisions"],"vendor_name":"Convai Innovations","vendor_icon":"https://avatars.githubusercontent.com/u/91363996?s=60\u0026v=4","billing_usage_schema":{"input_tokens":{"type":"number","unit":"token"}},"billing_usage_examples":[{"label":"Short request","facts":{"input_tokens":402}},{"label":"Maximum context","facts":{"input_tokens":65536}}],"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"32K","model_name":"jev-1.13.0","created_time":1789995214,"description":"Jev is a structured decision model from TypeSafe, and the first of its System One models. System One models make fast, structured decisions for software, returning a typed choice rather than free-form text. It is suited for routing, classification, and other decision points inside an application where a fast, predictable answer matters more than generated prose.\n\nLearn more in TypeSafe's docs: https://docs.typesafe.ai/concepts/system-one","vendor_id":46,"quota_type":0,"model_ratio":0.021,"model_price":0,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai","systemone"],"input_modalities":["text"],"output_modalities":["decisions"],"vendor_name":"TypeSafe","vendor_icon":"https://cdn.marmot-cloud.com/storage/zenmux/2026/09/20/BFx2T9Y/Property-1Typesafe.svg","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", u(\"input_tokens\") * 0.042 / 1000000)","billing_usage_schema":{"input_tokens":{"type":"number","unit":"token"}},"billing_usage_examples":[{"label":"Short request","facts":{"input_tokens":402}},{"label":"Maximum context","facts":{"input_tokens":65536}}],"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":0.5,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"1M","max_output_tokens_display":"128K","usage_tokens":88,"model_name":"glm-5.3-flashx","created_time":1789780562,"description":"GLM-5.3-FlashX is a faster and smoother version of GLM-5.3-Flash, delivering inference speeds of up to 200 tokens/s. GLM-5.3-FlashX is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while reducing compute overhead.","tags":"Chat","vendor_id":18,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file","video"],"output_modalities":["text"],"vendor_name":"Z.ai","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/acc3a4676ad771bd616079c832b3d7a5_1779342707107.svg","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 0.375 + c * 1.25 + cr * 0.075)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"1.05M","max_output_tokens_display":"1.05M","usage_tokens":380,"model_name":"kimi-k2.8-preview","created_time":1789780511,"description":"Kimi K2.8 Preview delivers overall performance close to K3 with more efficient thinking. It supports adjustable thinking effort levels aligned with K3’s reasoning tiers, and provides up to 1M-token ultra-long context.","tags":"Chat","vendor_id":25,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"Moonshot","vendor_icon":"Moonshot","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 1 + c * 4 + cr * 0.25)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":448,"model_name":"muse-spark-1.3","created_time":1789400437,"description":"Muse Spark 1.3 Contributor is the cost-efficient contributor tier of Meta’s multimodal reasoning model for experimentation, learning, and early-stage agentic, multi-agent, and coding workflows. It is designed to track information across extended tasks, work through conflicting inputs, and request clarification or confirmation when needed. Prompts and outputs may be used to improve Meta’s products.","tags":"Chat","vendor_id":45,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"vendor_name":"Meta","vendor_icon":"Meta","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 1.25 + c * 4.25 + cr * 0.15)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"400K","usage_tokens":7686,"model_name":"gpt-image-2.5-sunburst","created_time":1789041044,"description":"GPT Image 2.5 Sunburst is an image generation and editing model from OpenAI, positioned as the precision-oriented tier of the GPT Image 2.5 series. It is suited to detailed creative work where editing accuracy matters more than generation speed, via the dedicated Images API.","tags":"Image","vendor_id":5,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 5 + c * 30 + cr * 2 + img * 8)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":0.9,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"400K","usage_tokens":6835,"model_name":"gpt-image-2.5-flare","created_time":1789040974,"description":"GPT Image 2.5 Flare is an image generation and editing model from OpenAI, positioned as the speed-oriented tier of the GPT Image 2.5 series. It is suited to high-volume everyday generation, creator content, and rapid prototyping via the dedicated Images API.","tags":"Image","vendor_id":5,"quota_type":0,"model_ratio":4,"model_price":0,"owner_by":"","completion_ratio":2,"image_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 5 + c * 30 + cr * 2 + img * 8)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":0.9,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":544,"model_name":"deepseek-v4.1-flash","created_time":1789027793,"description":"DeepSeek V4.1 Flash is a 552B-parameter MoE model built on a new causal encoder-decoder architecture, designed for higher capability, faster reasoning, higher throughput, and lower serving cost. It natively supports multimodal visual understanding and delivers flagship-level intelligence with significantly reduced KV cache requirements, making it well suited for high-throughput and cost-sensitive agentic workloads.","tags":"Chat","vendor_id":16,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1","Tier 2","Tier 3"],"supported_endpoint_types":["openai"],"input_modalities":["text","image"],"output_modalities":["text"],"vendor_name":"DeepSeek","vendor_icon":"DeepSeek.Color","billing_mode":"tiered_expr","billing_expr":"(tier(\"base\", p * 0.3 + c * 1.2 + cr * 0.006)) * (hour(\"Asia/Shanghai\") \u003e= 9 \u0026\u0026 hour(\"Asia/Shanghai\") \u003c 12 \u0026\u0026 hour(\"Asia/Shanghai\") \u003e= 14 \u0026\u0026 hour(\"Asia/Shanghai\") \u003c 18 ? 0.5 : 1)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":0.5,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"1.05M","max_output_tokens_display":"128K","usage_tokens":493301714,"model_name":"gpt-6-astra","created_time":1788581859,"description":"GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon agentic tasks that involve computer and browser use.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":2,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","billing_mode":"tiered_expr","billing_expr":"len \u003c= 272000 ? tier(\"standard\", p * 10 + c * 50 + cr * 1 + cc * 12.5) : tier(\"long_context\", p * 20 + c * 75 + cr * 2 + cc * 25)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"262K","max_output_tokens_display":"32K","usage_tokens":14387707,"model_name":"ling-3.0-flash-vl","created_time":1788526087,"description":"Ling-3.0-flash-VL builds on Ling-3.0-flash, further strengthening its language capabilities while adding native visual perception and advanced visual agent capabilities.","tags":"Chat","vendor_id":43,"quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"InclusionAI","vendor_icon":"https://maas-public.oss-cn-shanghai.aliyuncs.com/maas-admin/image/2026-06-18/image-5b9c856a-3b04-40.png","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":0.1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":673,"model_name":"ling-3.0-flash-sante","created_time":1788526051,"description":"Ling-3.0-Flash-Sante is a Mixture-of-Experts (MoE) model with enhanced capabilities across health and medicine. Built on Ling-3.0-Flash, it has 124 billion total parameters and activates approximately 5.1 billion parameters per token. The model excels in medical knowledge reasoning, clinical safety, evidence-based retrieval, and long-horizon medical tasks, while retaining strong general capabilities in reasoning, coding, and agentic tasks.","tags":"Chat","vendor_id":43,"quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"","completion_ratio":1,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"InclusionAI","vendor_icon":"https://maas-public.oss-cn-shanghai.aliyuncs.com/maas-admin/image/2026-06-18/image-5b9c856a-3b04-40.png","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":0.1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":733,"model_name":"ling-3.0-flash-fin","created_time":1788525976,"description":"Ling-3.0-flash-Fin is a finance-enhanced MoE model built on Ling-3.0-Flash, with 124 billion total parameters and approximately 5.1 billion activated parameters. Designed for real-world investment workflows, it is optimized for complex multi-step tasks and long-horizon planning and execution. With a relatively small active parameter footprint, it delivers competitive financial performance while maintaining strong general capabilities in reasoning, coding, and mathematics.","tags":"Chat","vendor_id":43,"quota_type":0,"model_ratio":0.03,"model_price":0,"owner_by":"","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"InclusionAI","vendor_icon":"https://maas-public.oss-cn-shanghai.aliyuncs.com/maas-admin/image/2026-06-18/image-5b9c856a-3b04-40.png","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":0.1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Hailuo-H3-Max","created_time":1788517553,"description":"MiniMax-H3-Max is a high-speed video generation model post-trained on MiniMax H3, with enhanced prompt-adherence tuning to achieve “real-time” video generation—where generation speed exceeds playback speed—enabling 24/7 AI live streaming and real-time interactive content creation.","tags":"Video","vendor_id":10,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 1","Tier 2","Tier 3","Tier 4","default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"768p","multiplier":8},{"value":"1080p","multiplier":10},{"value":"2048p","multiplier":15},{"value":"4096p","multiplier":18}]}},{"usage_tokens":71,"model_name":"gemini-3.8-flash","created_time":1788415055,"description":"Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning.","tags":"Chat","vendor_id":6,"quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":0.5,"group-free":0,"svip":1,"vip":1}},{"model_name":"Vega-1.0-pro","created_time":1788371700,"tags":"Video","vendor_id":42,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image","video"],"output_modalities":["video"],"vendor_name":"Vega","vendor_icon":"Vega","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":15},{"value":"1080p","multiplier":30},{"value":"2048p","multiplier":72},{"value":"4096p","multiplier":86}]}},{"model_name":"Vega-1.0-lite","created_time":1788371684,"tags":"Video","vendor_id":42,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image","video"],"output_modalities":["video"],"vendor_name":"Vega","vendor_icon":"Vega","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":15},{"value":"1080p","multiplier":18},{"value":"2048p","multiplier":22},{"value":"4096p","multiplier":0.26}]}},{"usage_tokens":394,"model_name":"claude-fable-5-1","created_time":1788336565,"description":"Claude Fable 5.1 is Anthropic’s high-performance model designed for complex coding and knowledge work. It excels at long-running, multi-step tasks, with strong capabilities in autonomous planning, tool use, and problem-solving across agentic workflows, software development, deep research, and complex enterprise use cases.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 10 + c * 50 + cr * 0.25 + cc * 12.5 + cc1h * 20)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Wan-3.0-prime","created_time":1788234883,"description":"Wan3.0-Video-Prime is the speed-optimized version of the Wan video generation model, greatly improving generation speed while maintaining high-quality video output. It supports multi-modal inputs including text, image, video, and audio, unifying text-to-video, image-to-video (first frame/first-last frame), and reference-based video generation.","tags":"Video","vendor_id":28,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image","audio","video"],"output_modalities":["video"],"vendor_name":"Qwen","vendor_icon":"Qwen","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":14},{"value":"1080p","multiplier":28},{"value":"2048p","multiplier":35},{"value":"4096p","multiplier":40}]}},{"model_name":"Wan-3.0","created_time":1788234738,"description":"Wan3.0-Video-Prime is the speed-optimized version of the Wan video generation model, greatly improving generation speed while maintaining high-quality video output. It supports multi-modal inputs including text, image, video, and audio, unifying text-to-video, image-to-video (first frame/first-last frame), and reference-based video generation. ","tags":"Video","vendor_id":28,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image","audio","video"],"output_modalities":["video"],"vendor_name":"Qwen","vendor_icon":"Qwen","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":10},{"value":"1080p","multiplier":20},{"value":"2048p","multiplier":25},{"value":"4096p","multiplier":0.3}]}},{"usage_tokens":1413,"model_name":"hy4-preview","created_time":1787903811,"description":"Hy4 preview has 770B total parameters with 49B activated, supports a 1M-token context window, and stays compact within the mainstream flagship class — a single 8-GPU server hosts a full inference node. The model is optimized for Agent, Coding, and productivity scenarios, strengthening comprehension, planning, tool use, and sustained execution on complex tasks.Compared with its predecessor, Hy4 preview advances multi-step agent, coding, and productivity workloads, with more reliable task decomposition, context tracking, instruction following, and long-chain execution. For coding, it further improves code comprehension, generation, modification, and complex engineering; for productivity, it strengthens document processing, information analysis, office automation, game development, web page generation, and cross-tool collaboration.Hy4 preview is suited for coding agents, complex tool use, and agentic workflows that demand multi-step planning and sustained execution, delivering dependable task completion on complex, real-world business tasks.","tags":"Chat","vendor_id":2,"quota_type":0,"model_ratio":0.425,"model_price":0,"owner_by":"","completion_ratio":3,"cache_ratio":0.052941176471,"enable_groups":["default","Tier 1","Tier 2","Tier 3","TokenPlan-Official"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"Tencent","vendor_icon":"Hunyuan.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1723,"model_name":"qwen3.8-flash","created_time":1787881056,"description":"Qwen3.8-Flash is the latest multimodal large model launched by Qianwen, which combines powerful understanding and generation capabilities with excellent response speed. The model natively supports millions of contextual windows and can handle extremely long documents, code repositories, and complex conversations at once. It performs particularly well in scenarios such as programming assistance, intelligent agent collaboration, and graphic and text understanding - whether it's automatically fixing code, operating desktop applications, or analyzing charts and long videos, it can provide accurate and high-quality results.","tags":"Chat","vendor_id":28,"quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"","completion_ratio":2.666666666667,"cache_ratio":0.1,"create_cache_ratio":1.333333333333,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","audio","video"],"output_modalities":["text"],"vendor_name":"Qwen","vendor_icon":"Qwen","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1834,"model_name":"glm-5.3-flash","created_time":1787825918,"description":"GLM-5.3-Flash is the first natively multimodal model in the GLM-5 series, achieving stronger intelligence beyond GLM-5.2 through an extremely low-cost architecture. With 320B total parameters and 18B activated parameters, it is the first open-source frontier model to adopt a hybrid architecture combining sparse and linear attention. Visual capabilities are natively integrated into the coding loop, enabling the model to proactively observe interfaces, rendering results, and interactive feedback, and continuously test and improve accordingly.","tags":"Chat","vendor_id":18,"quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"","completion_ratio":3.333333333333,"cache_ratio":0.2,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"Z.ai","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/acc3a4676ad771bd616079c832b3d7a5_1779342707107.svg","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1723,"model_name":"qwen3.8-max","created_time":1787728300,"description":"2.4-trillion-parameter MoE flagship delivering a comprehensive leap in coding and professional work. Autonomously codes and delivers complete projects spanning 10+ days. Handles hundreds of specialized tasks across legal, financial, design, and other professional domains, producing production-grade results end-to-end in a single conversation. Native visual understanding runs through the full cycle of planning, execution, and verification, enabling deep semantic analysis of ultra-long documents and extended video content. In long-horizon tasks, plans autonomously, iterates through closed feedback loops, and continuously evolves.","tags":"Chat","vendor_id":28,"quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"","completion_ratio":2.942857142857,"cache_ratio":0.142857142857,"create_cache_ratio":1.228571428571,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"Qwen","vendor_icon":"Qwen","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":3657,"model_name":"deepseek-v4-flash-vision-exp","created_time":1787431922,"description":"DeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of DeepSeek V4 Flash 0731(opens in new tab) from DeepSeek, adding image understanding while matching the base model on text capabilities including agents, reasoning, and world knowledge. It is a sparse mixture-of-experts model with 13B active parameters out of 284B total.\n\nIt is suited for document and chart understanding, visual question answering, and multimodal agent workflows that interleave text and images.","vendor_id":16,"quota_type":0,"model_ratio":0.225,"model_price":0,"owner_by":"","completion_ratio":2.888888888889,"cache_ratio":0.333333333333,"enable_groups":["default","Tier 1","Tier 2","Tier 3"],"supported_endpoint_types":["openai"],"input_modalities":["text","image"],"output_modalities":["text"],"vendor_name":"DeepSeek","vendor_icon":"DeepSeek.Color","billing_mode":"tiered_expr","billing_expr":"(tier(\"base\", p * 0.45 + c * 1.3 + cr * 0.015)) * ((hour(\"Asia/Shanghai\") \u003e= 9 || hour(\"Asia/Shanghai\") \u003c 12) \u0026\u0026 (hour(\"Asia/Shanghai\") \u003e= 14 || hour(\"Asia/Shanghai\") \u003c 18) ? 0.5 : 1)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"doubao-seedance-2-5-260628","created_time":1786701437,"description":"Seedance 2.5 is the next-generation multimodal video creation model developed by the Doubao foundation model team, ushering in a new era of industrial-grade video creation with **long-form storytelling, multi-asset support, powerful editing capabilities, and multilingual support.","tags":"Video","vendor_id":17,"quota_type":0,"model_ratio":5.85,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai-video","openai"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["video"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_usage_schema":{"resolution":{"enum":["480p","720p","1080p"],"description":{"en":"Output video resolution","zh":"输出视频分辨率"},"enumLabels":{"1080p":{"en":"1080p","zh":"1080p"},"480p":{"en":"480p","zh":"480p"},"720p":{"en":"720p","zh":"720p"}}},"tokens":{"type":"number","unit":"token","description":{"en":"Billing token unit price","zh":"计费 Token 单价"}},"video_input":{"enum":["none","video"],"description":{"en":"Reference video input","zh":"参考视频输入"},"enumLabels":{"none":{"en":"No reference video","zh":"无参考视频"},"video":{"en":"With reference video","zh":"有参考视频"}}}},"billing_usage_examples":[{"label":"480p · 5s","facts":{"resolution":"480p","tokens":48038,"video_input":"none"}},{"label":"720p · 5s","facts":{"resolution":"720p","tokens":108000,"video_input":"none"}},{"label":"1080p · 5s","facts":{"resolution":"1080p","tokens":243000,"video_input":"none"}},{"label":"720p · 5s (+4s 输入视频)","facts":{"resolution":"720p","tokens":194400,"video_input":"video"}}],"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":459,"model_name":"gemini-3.7-flash","created_time":1786691990,"description":"Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. It is designed for tasks that require responsive performance and reliable multi-step problem solving.","tags":"Chat","vendor_id":6,"quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":0.5,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1435,"model_name":"glm-5.3","created_time":1786688556,"description":"Zhipu's latest flagship model with 1M context window, supporting controllable thinking length","tags":"Chat","vendor_id":18,"quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"","completion_ratio":3,"cache_ratio":0.2,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"Z.ai","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/acc3a4676ad771bd616079c832b3d7a5_1779342707107.svg","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":271266,"model_name":"grok-4.6","created_time":1786617785,"description":"Grok 4.6 is SpaceXAI's smartest model with frontier performance on coding, knowledge work, and STEM.","tags":"Chat","vendor_id":14,"quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image"],"output_modalities":["text"],"vendor_name":"xAI","vendor_icon":"https://t0.gstatic.com/faviconV2?client=SOCIAL\u0026type=FAVICON\u0026fallback_opts=TYPE,SIZE,URL\u0026url=https://x.ai/\u0026size=256","billing_mode":"tiered_expr","billing_expr":"len \u003c= 200000 ? tier(\"standard\", p * 2 + c * 6 + cr * 0.5) : tier(\"long_context\", p * 4 + c * 12 + cr * 1)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":5002,"model_name":"gemini-3.5-flash-lite","created_time":1786462713,"description":"The fastest, most cost-effective 3.5-class model, delivering 350 output tokens per second . It offers flexible thinking levels and excels in low-latency and high-throughput agentic tasks.","tags":"Chat","vendor_id":6,"quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"","completion_ratio":8.333333333333334,"cache_ratio":0.1,"create_cache_ratio":0.000003333333,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"minimaxmusic-3.0","created_time":1785813595,"tags":"Music","vendor_id":10,"quota_type":1,"model_ratio":0,"model_price":0.2,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["audio"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1,"model_name":"MJ:v8.2","created_time":1785764652,"tags":"Image","vendor_id":15,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 1","default"],"supported_endpoint_types":["openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Midjourney","vendor_icon":"Midjourney","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 68000)\n  : param(\"resolution\") == \"2K\"\n    ? tier(\"2K\", 75000)\n    : param(\"resolution\") == \"4K\"\n      ? tier(\"4K\", 80000)\n      : tier(\"default\", 80000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"MJ:niji_7","created_time":1785665637,"tags":"Image","vendor_id":15,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 1","default"],"supported_endpoint_types":["openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Midjourney","vendor_icon":"Midjourney","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 68000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 75000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 80000)\n  : tier(\"default\", 80000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Nano-banana2-lite","created_time":1785665119,"description":" (Nano Banana Lite) is designed as the efficiency specialist of the image generation family, offering ultra-low latency and cost-effective image generation and editing. By targeting a sub-2 second latency and significantly reduced TPU compute costs, this model enables high-volume interactive developer use cases and real-time consumer applications.","tags":"Image","vendor_id":6,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Google","vendor_icon":"Gemini.Color","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 35000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 42000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 50000)\n  : tier(\"default\", 50000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":72808,"model_name":"claude-opus-5","created_time":1784978856,"description":"Claude Opus 5 is Anthropic’s flagship model for complex agentic coding and enterprise workloads. Compared with the previous generation, it delivers significant improvements in deep reasoning, long-horizon task execution, multi-agent collaboration, and complex professional work, enabling it to more reliably handle demanding tasks that require sustained planning, tool use, and multi-step reasoning.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 5 + c * 25 + cr * 0.5 + cc * 6.25 + cc1h * 10)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":151,"model_name":"gemini-3.6-flash","created_time":1784795948,"description":"Gemini 3.6 Flash is a high-efficiency model from Google for coding, agentic workflows, and web and app development. It is designed to produce polished outputs with fewer unnecessary edits and less hedging, while reducing token use and the number of model calls needed to complete a task.","tags":"Chat","vendor_id":6,"quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":6.66667e-7,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":245181,"model_name":"kimi-k3","created_time":1784279236,"description":"Kimi K3 is Kimi’s most capable model to date, with 2.8 trillion parameters. Built on Kimi Delta Attention, a hybrid linear attention mechanism, and Attention Residuals, it offers native visual understanding and a 1M-token context window for frontier intelligence scenarios such as software engineering, knowledge work, and deep reasoning.","tags":"Chat","vendor_id":27,"quota_type":0,"model_ratio":1.43,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"Kimi","vendor_icon":"Kimi","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":192,"model_name":"gemini-3.1-flash-lite","created_time":1783763999,"description":"Gemini 3.1 Flash-Lite is Google's most cost-effective model, designed for high-volume, cost-sensitive LLM traffic and optimized for low-latency use cases.","tags":"Chat","vendor_id":6,"quota_type":0,"model_ratio":0.0125,"model_price":0,"owner_by":"","completion_ratio":60,"cache_ratio":0.00004,"image_ratio":10,"audio_ratio":20,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1609205,"model_name":"gpt-5.6-terra","created_time":1783659746,"description":"GPT-5.6 Terra is a balanced model in OpenAI's GPT-5.6 series, positioned between the flagship Sol tier and the cost-efficient Luna tier. It is suited for everyday coding, reasoning, and agentic tasks where capability and cost need to be balanced, offering strong performance at roughly half the cost of Sol.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"","completion_ratio":6,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","billing_mode":"tiered_expr","billing_expr":"len \u003c= 272000 ? tier(\"standard\", p * 2 + c * 12 + cr * 0.2 + cc * 2.5) : tier(\"long_context\", p * 4 + c * 18 + cr * 0.4 + cc * 5)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"context_length_display":"1.05M","max_output_tokens_display":"128K","usage_tokens":105051489,"model_name":"gpt-5.6-sol","created_time":1783659645,"description":"GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly strong at command-line and multi-step coding tasks and long-horizon problem solving.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"","completion_ratio":6,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","billing_mode":"tiered_expr","billing_expr":"len \u003c= 272000 ? tier(\"standard\", p * 5 + c * 30 + cr * 0.5 + cc * 6.25 + cc1h * 6.25) : tier(\"long_context\", p * 10 + c * 45 + cr * 1 + cc * 12.5 + cc1h * 12.5)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":23321921,"model_name":"gpt-5.6-luna","created_time":1783659505,"description":"GPT-5.6 Luna is a fast, cost-efficient model in OpenAI's GPT-5.6 series. It is suited for high-volume, latency-sensitive tasks such as chat, classification, and lightweight agentic workflows, providing capable reasoning for its price tier.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":0.5,"model_price":0,"owner_by":"","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","billing_mode":"tiered_expr","billing_expr":"len \u003c= 272000 ? tier(\"standard\", p * 0.2 + c * 1.2 + cr * 0.02 + cc * 0.25) : tier(\"long_context\", p * 0.4 + c * 1.8 + cr * 0.04 + cc * 0.5)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":88934,"model_name":"grok-4.5","created_time":1783659310,"description":"Grok 4.5 is SpaceXAI's smartest model with frontier performance on coding, knowledge work, and STEM.","tags":"Chat","vendor_id":14,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"xAI","vendor_icon":"https://t0.gstatic.com/faviconV2?client=SOCIAL\u0026type=FAVICON\u0026fallback_opts=TYPE,SIZE,URL\u0026url=https://x.ai/\u0026size=256","billing_mode":"tiered_expr","billing_expr":"len \u003c= 200000 ? tier(\"standard\", p * 2 + c * 6 + cr * 0.5) : tier(\"long_context\", p * 4 + c * 12 + cr * 1)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"gpt-image-2","created_time":1783593172,"description":"GPT-image-2 is OpenAI’s latest image generation model, designed as a production-grade engine for visual content creation. It supports text-to-image, image-to-image, and inpainting, with significant improvements in image quality, detail fidelity, and instruction following. The model offers enhanced control over resolution and aspect ratios, and strengthens multilingual understanding and real-world semantic representation. It can generate more consistent and controllable visual outputs, and supports fine-grained editing of existing images (such as localized replacement and style adjustments). Overall, it targets e-commerce, marketing, and design scenarios, making it well-suited for large-scale content production.","tags":"Image","vendor_id":5,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":2,"enable_groups":["default"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 5 + c * 30 + img * 8)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"grok-imagine-image-quality","created_time":1783482785,"description":"grok-imagine-image-quality is xAI’s high-quality image generation model. It offers the same feature set as the standard tier ,suitable for professional design, marketing materials, and other applications requiring high image quality.","tags":"Image","vendor_id":14,"quota_type":1,"model_ratio":0,"model_price":0.075,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"xAI","vendor_icon":"https://t0.gstatic.com/faviconV2?client=SOCIAL\u0026type=FAVICON\u0026fallback_opts=TYPE,SIZE,URL\u0026url=https://x.ai/\u0026size=256","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":2,"model_name":"grok-imagine-image","created_time":1783482727,"description":"grok-imagine-image is the standard image generation model developed by xAI. Capable of generating images from text prompts, this model boasts extensive configuration features including batch generation, aspect ratio adjustment and resolution customization. Delivering outstanding cost performance, it can accommodate large-scale image generation workloads.","tags":"Image","vendor_id":14,"quota_type":1,"model_ratio":0,"model_price":0.025,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"xAI","vendor_icon":"https://t0.gstatic.com/faviconV2?client=SOCIAL\u0026type=FAVICON\u0026fallback_opts=TYPE,SIZE,URL\u0026url=https://x.ai/\u0026size=256","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1510,"model_name":"kimi-k2.7-code-highspeed","created_time":1783420594,"description":"Kimi K2.7 Code HighSpeed is the high-speed version of Kimi K2.7 Code, the same model as Kimi K2.7 Code, but with an output speed of approximately 180 Tokens/s and up to 260 Tokens/s in short context scenarios, delivering a more extreme coding experience.","tags":"Chat","vendor_id":27,"quota_type":0,"model_ratio":0.9295,"model_price":0,"owner_by":"","completion_ratio":4.153846153846,"cache_ratio":0.2,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"Kimi","vendor_icon":"Kimi","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":2514,"model_name":"hy3","created_time":1783383290,"description":"Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort: a direct no-think mode by default, plus low and high chain-of-thought modes for complex math, coding, and multi-step problems. With a 256K context window, Hy3 targets long-horizon tasks, including improved coreference resolution, multi-turn constraint tracking, and stable tool-calling that generalizes across agent scaffoldings.\n\nTencent positions it as a reliable, cost-effective option across coding, document processing, financial analysis, game development, and frontend design, with a strong emphasis on grounded, anti-hallucination behavior that answers when grounded and flags when evidence is missing rather than fabricating.","tags":"Chat","vendor_id":2,"quota_type":0,"model_ratio":0.0735,"model_price":0,"owner_by":"","completion_ratio":4.006802721088,"cache_ratio":0.251700680272,"enable_groups":["default","Tier 1","Tier 2","Tier 3","TokenPlan-Official"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"Tencent","vendor_icon":"Hunyuan.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"gemini-3.1-flash-lite-image","created_time":1783030979,"description":"Nano Banana Lite is designed as the efficiency specialist of the image generation family, offering ultra-low latency and cost-effective image generation and editing. By targeting a sub-2 second latency and significantly reduced TPU compute costs, this model enables high-volume interactive developer use cases and real-time consumer applications.","tags":"Image","vendor_id":6,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Google","vendor_icon":"Gemini.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 0.25 + c * 1.5 + img_o * 30 + ao * 30)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":150741,"model_name":"claude-sonnet-5","created_time":1782887103,"description":"Claude Sonnet 5 is Anthropic’s latest-generation Sonnet series model, positioned as the most agent-capable Sonnet model. It demonstrates significant improvements over previous versions in tool use, code generation, and multi-step task execution. The model supports browser and terminal tool calling, enabling autonomous planning and execution of complex tasks. Across multiple benchmarks, its performance approaches Claude Opus 4.8 while maintaining lower operational costs, making it well-suited for cost-efficient agentic systems and enterprise applications.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 3 + c * 15 + cr * 0.3 + cc * 3.75 + cc1h * 6)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":7441,"model_name":"minimax-m2.7-highspeed","created_time":1782540127,"description":"MiniMax-M2.7-highspeed is the MiniMax-M2.7 High-Speed edition — same performance, faster and more agile (output speed about 100 TPS).","tags":"Chat","vendor_id":10,"quota_type":0,"model_ratio":0.294,"model_price":0,"owner_by":"","completion_ratio":4,"cache_ratio":0.1,"create_cache_ratio":0.625,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":7441,"model_name":"minimax-m2.5-lightning","created_time":1782540072,"description":"MiniMax-M2.5-lightning is a new-generation text and agent model launched by MiniMax. As a major upgrade to the M2 series, it focuses on enhancing programming capabilities, tool invocation, and multi-step Agent workflow execution, targeting complex automation and long-task scenarios. Compared to MiniMax-M2.5, MiniMax-M2.5-lightning delivers the same overall output quality, but with faster and more agile responses.","tags":"Chat","vendor_id":10,"quota_type":0,"model_ratio":0.147,"model_price":0,"owner_by":"","completion_ratio":8,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Happyhorse-1.1","created_time":1782401770,"description":"HappyHorse-1.1 generation with improved visual quality, dynamic performance, and cross-clip consistency. It more accurately understands the input image and preserves creative intent, delivering significant improvements in skin texture realism, character ID consistency across clips, motion smoothness, text rendering stability, and audio-visual synchronization, producing high-quality videos with greater realism, richer details, and stronger overall consistency.","tags":"Video","vendor_id":28,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Qwen","vendor_icon":"Qwen","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":14},{"value":"1080p","multiplier":20},{"value":"2048p","multiplier":32},{"value":"4096p","multiplier":38}]}},{"model_name":"MJ:v8.1","created_time":1782401612,"tags":"Image","vendor_id":15,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Midjourney","vendor_icon":"Midjourney","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 68000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 75000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 80000)\n  : tier(\"default\", 80000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"doubao-seedance-2-0-mini-260615","created_time":1782320965,"description":"Seedance 2.0 mini is a new generation of cost-effective video generation model launched to meet a wider range of video generation needs. While maintaining competitive effects, it brings video generation capabilities to application scenarios with lower thresholds, higher frequencies, and greater scalability","tags":"Video,Async","vendor_id":17,"quota_type":0,"model_ratio":1.75,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai-video","openai"],"input_modalities":["text","image","audio","video"],"output_modalities":["video"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_usage_schema":{"resolution":{"enum":["480p","720p"],"description":{"en":"Output video resolution","zh":"输出视频分辨率"},"enumLabels":{"480p":{"en":"480p","zh":"480p"},"720p":{"en":"720p","zh":"720p"}}},"tokens":{"type":"number","unit":"token","description":{"en":"Billing token unit price","zh":"计费 Token 单价"}},"video_input":{"enum":["none","video"],"description":{"en":"Reference video input","zh":"参考视频输入"},"enumLabels":{"none":{"en":"No reference video","zh":"无参考视频"},"video":{"en":"With reference video","zh":"有参考视频"}}}},"billing_usage_examples":[{"label":"480p · 5s","facts":{"resolution":"480p","tokens":48038,"video_input":"none"}},{"label":"720p · 5s","facts":{"resolution":"720p","tokens":108000,"video_input":"none"}},{"label":"720p · 5s (+4s 输入视频)","facts":{"resolution":"720p","tokens":194400,"video_input":"video"}}],"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"doubao-seedream-5-0-260128","created_time":1782222424,"description":"Dola-Seedream-5.0-lite is the latest image generation model released by BytePlus. For the first time, it introduces web-connected retrieval, enabling the model to fuse real-time online information to significantly improve the timeliness and relevance of generated images. The model’s reasoning and comprehension capabilities are further upgraded, allowing it to accurately interpret complex prompts and visual inputs. In addition, Dola-Seedream-5.0-lite delivers notable improvements in global knowledge coverage, reference consistency, and professional-grade scene generation, making it well suited for enterprise-level visual creation workflows.\n","tags":"Image","vendor_id":17,"quota_type":1,"model_ratio":0,"model_price":0.035,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_usage_schema":{"images_above_1_5k":{"type":"number","unit":"count","unitLabel":{"en":"image","fr":"image","ja":"枚","ru":"изображение","vi":"ảnh","zh":"张","zh-TW":"張"},"description":{"en":"Image generation unit price (above 1.5K)","zh":"图片生成单价（1.5K 以上）"}},"images_up_to_1_5k":{"type":"number","unit":"count","unitLabel":{"en":"image","fr":"image","ja":"枚","ru":"изображение","vi":"ảnh","zh":"张","zh-TW":"張"},"description":{"en":"Image generation unit price (1.5K and below)","zh":"图片生成单价（1.5K 及以下）"}},"input_images":{"type":"number","unit":"count","unitLabel":{"en":"image","fr":"image","ja":"枚","ru":"изображение","vi":"ảnh","zh":"张","zh-TW":"張"},"description":{"en":"Input image unit price","zh":"输入图片单价"}},"layer_decomposition":{"type":"boolean","description":{"en":"Whether layer decomposition is enabled","zh":"是否开启图层拆分"}}},"billing_usage_examples":[{"label":"2K · 1 张","facts":{"images_above_1_5k":1,"images_up_to_1_5k":0,"input_images":0,"layer_decomposition":false}},{"label":"1K · 1 张 · 2 张参考图","facts":{"images_above_1_5k":0,"images_up_to_1_5k":1,"input_images":2,"layer_decomposition":false}},{"label":"2K · 4 张组图","facts":{"images_above_1_5k":4,"images_up_to_1_5k":0,"input_images":0,"layer_decomposition":false}},{"label":"图层拆分 · 2K 底图 + 4 层 1.5K","facts":{"images_above_1_5k":1,"images_up_to_1_5k":4,"input_images":1,"layer_decomposition":true}}],"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":8773,"model_name":"doubao-seed-evolving","created_time":1782222328,"description":"Doubao-Seed-Evolving is a dynamically iterated model of the Seed series designed for Agent and Coding scenarios, featuring core capabilities including complex task orchestration, long-range planning, code generation, and tool calling. The model adopts a dynamic iteration mechanism , ensuring continuous capability improvement.","tags":"Chat","vendor_id":17,"quota_type":0,"model_ratio":0.42,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":8621,"model_name":"doubao-seed-2-1-turbo-260628","created_time":1782222160,"description":"doubao-seed-2-1-turbo-260628 is the low-cost, low-latency version of the Doubao Seed 2.1 series designed for scaled production scenarios. It is suitable for enterprise deployment scenarios that require stable handling of high-volume online calls for Coding, Agent, and multimodal understanding tasks, where cost, throughput, and batch invocation are priorities.","tags":"Chat","vendor_id":17,"quota_type":0,"model_ratio":0.21,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":8573,"model_name":"doubao-seed-2-1-pro-260628","created_time":1782222088,"description":"doubao-seed-2-1-pro-260628 is the latest flagship model of ByteDance’s Doubao series, designed for the Coding and Agent era, with comprehensive upgrades in coding engineering delivery, long-chain Agent task execution, and multimodal understanding. The model demonstrates stronger requirement comprehension, long-term planning, and continuous repair capabilities.","tags":"Chat","vendor_id":17,"quota_type":0,"model_ratio":0.42,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.2,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":43598,"model_name":"mimo-v2.5","created_time":1781865464,"description":"MiMo-V2.5 is a native omnimodal model by Xiaomi. It delivers Pro-level agentic performance at roughly half the inference cost, while surpassing MiMo-V2-Omni in multimodal perception across image and video understanding tasks. Its 1M context window supports complete documents, extended conversations, and complex task contexts in a single pass, making it ideal for integration with agent frameworks where strong reasoning, rich perception, and cost efficiency all matter.","tags":"Chat","vendor_id":41,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":2,"cache_ratio":0.003,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","audio","video"],"output_modalities":["text"],"vendor_name":"Xiaomi","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/80caddd717cfafd45b0671c5b8969296_1779343536965.svg","billing_mode":"tiered_expr","billing_expr":"len \u003c= 256000 ? tier(\"standard\", p * 0.4 + c * 2 + cr * 0.08) : tier(\"long_context\", p * 0.8 + c * 4 + cr * 0.16)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":44554,"model_name":"mimo-v2.5-pro","created_time":1781865367,"description":"MiMo-V2.5-Pro is Xiaomi’s flagship model, delivering strong performance in general agentic capabilities, complex software engineering, and long-horizon tasks, with top rankings on benchmarks such as ClawEval, GDPVal, and SWE-bench Pro. It can independently and autonomously complete professional tasks that would take human experts days or weeks, involving more than a thousand tool calls. Its context length of up to 1M makes it well suited for integration with a wide range of agent frameworks.","tags":"Chat","vendor_id":41,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":2,"cache_ratio":0.0045,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"Xiaomi","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/80caddd717cfafd45b0671c5b8969296_1779343536965.svg","billing_mode":"tiered_expr","billing_expr":"len \u003c= 256000 ? tier(\"standard\", p * 1 + c * 3 + cr * 0.2) : tier(\"long_context\", p * 2 + c * 6 + cr * 0.4)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":4635,"model_name":"qwen3.7-plus","created_time":1781796148,"description":"Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series. It supports text and image input with text output, building on the series' text capabilities with a comprehensive upgrade to its vision-language abilities while retaining full-stack, agent-level intelligence for coding, tool use, and productivity workflows. Its distinguishing trait is multi-modal interactive hybrid agent capability: it can perceive real-world scenes, read screens and interact with GUIs, generate code from visual references, and perform end-to-end navigation within mobile apps.","tags":"Chat","vendor_id":28,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":4,"cache_ratio":0.064,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image"],"output_modalities":["text"],"vendor_name":"Qwen","vendor_icon":"Qwen","billing_mode":"tiered_expr","billing_expr":"len \u003c= 256000 ? tier(\"standard\", p * 0.3 + c * 1.15 + cr * 0.03 + cc * 0.35) : tier(\"long_context\", p * 0.85 + c * 3.4 + cr * 0.085 + cc * 1.05)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":31989,"model_name":"minimax-m3","created_time":1781796063,"description":"MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding, and tool use. It is built on MiniMax Sparse Attention (MSA), which replaces full attention with KV-block selection to cut per-token compute at long context — roughly 1/20 the cost of the previous generation at 1M tokens, with substantially faster prefill and decode while retaining quality across most tasks.\n\nTrained as a native multimodal model on interleaved data and tuned for multi-turn, production-like collaboration via an interactive user-simulator framework, the model is oriented toward sustained, multi-step tasks rather than single-turn execution.","tags":"Chat","vendor_id":10,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":4,"cache_ratio":0.06,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"MiniMax","vendor_icon":"Minimax","billing_mode":"tiered_expr","billing_expr":"len \u003c= 512000 ? tier(\"standard\", p * 0.6 + c * 2.4 + cr * 0.12) : tier(\"long_context\", p * 1.2 + c * 4.8 + cr * 0.24)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1426,"model_name":"gemini-3-pro-image","created_time":1781795682,"description":"Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro. It extends the original Nano Banana with significantly improved multimodal reasoning, real-world grounding, and high-fidelity visual synthesis. The model generates context-rich graphics, from infographics and diagrams to cinematic composites, and can incorporate real-time information via Search grounding.\n\nIt offers industry-leading text rendering in images (including long passages and multilingual layouts), consistent multi-image blending, and accurate identity preservation across up to five subjects. Nano Banana Pro adds fine-grained creative controls such as localized edits, lighting and focus adjustments, camera transformations, and support for 2K/4K outputs and flexible aspect ratios. It is designed for professional-grade design, product visualization, storyboarding, and complex multi-element compositions while remaining efficient for general image creation workflows.","tags":"Image","vendor_id":6,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image"],"output_modalities":["text","image"],"vendor_name":"Google","vendor_icon":"Gemini.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 2 + c * 12 + img_o * 120)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1919,"model_name":"kimi-k2.7-code","created_time":1781795416,"description":"MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts. It uses a native multimodal mixture-of-experts architecture that accepts text and image input, and it always operates in a thinking mode, preserving full reasoning content across multi-turn conversations. With a 256K-token context window, it targets long-horizon coding, agentic task decomposition, and multi-turn dialogue. The model activates 32B parameters out of roughly 1T total.","tags":"Chat","vendor_id":27,"quota_type":0,"model_ratio":0.455,"model_price":0,"owner_by":"","completion_ratio":4.153846153846,"cache_ratio":0.2,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"Kimi","vendor_icon":"Kimi","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":2578,"model_name":"glm-5.2","created_time":1781795292,"description":"GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering, and complex multi-step automation.\n\nReasoning efforts high and xhigh are supported; xhigh maps to max reasoning. It is particularly strong at coding and tool use across long-running tasks, able to maintain engineering context and follow standards consistently through a full development workflow, from requirements to multi-platform deployment, in a single task.","tags":"Chat","vendor_id":18,"quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"","completion_ratio":3.2,"cache_ratio":0.2,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"Z.ai","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/acc3a4676ad771bd616079c832b3d7a5_1779342707107.svg","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Vidu-q3-drama","created_time":1781794708,"tags":"Video","vendor_id":1,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Vidu","vendor_icon":"Vidu","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"1080p","multiplier":15},{"value":"2048p","multiplier":20},{"value":"4096p","multiplier":25}],"audio_generation_surcharge":0.08}},{"model_name":"Kling-3.0-turbo","created_time":1781794530,"tags":"Video","vendor_id":4,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 4","default","Tier 1","Tier 2","Tier 3"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Kwai","vendor_icon":"Kling.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":13},{"value":"1080p","multiplier":18},{"value":"2048p","multiplier":20},{"value":"4096p","multiplier":25}]}},{"usage_tokens":132,"model_name":"nex-n2-pro","created_time":1781211424,"description":"Nex-N2-Pro is an agentic mixture-of-experts model from Nex AGI, with 17B active parameters out of 397B total. Built on the Qwen3.5 architecture, it accepts text and image input and produces text output, and supports reasoning, function calling, and structured outputs. It is designed for coding, tool use, deep research, and long-horizon agentic workflows, unifying planning, code implementation, debugging, and iteration into a single execution loop.","tags":"Chat","vendor_id":40,"quota_type":0,"model_ratio":0.25,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"上海创智学院","vendor_icon":"https://sf-maas.s3.us-east-1.amazonaws.com/Model_LOGO/NEX.svg","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1433918,"model_name":"claude-fable-5","created_time":1781070678,"description":"Claude Fable 5 is a Mythos-class model from Anthropic, built for autonomous knowledge work and coding. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token context window. It is suited for long-running, complex, and asynchronous tasks that previously required frequent human check-ins.\n\nIt is particularly strong at end-to-end work that would otherwise take a person hours, days, or weeks - taking on problems that are long-running, ambiguous, or highly multi-step. It executes well-scoped tasks with few mistakes, automatically self-correcting through verification loops, and ships with robust safeguards.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 10 + c * 50 + cr * 1 + cc * 12.5 + cc1h * 20)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"gemini-3.1-flash-image","created_time":1780786695,"description":"Gemini 3.1 Flash Image Preview, a.k.a. \"Nano Banana 2,\" is Google’s latest state of the art image generation and editing model, delivering Pro-level visual quality at Flash speed. It combines advanced contextual understanding with fast, cost-efficient inference, making complex image generation and iterative edits significantly more accessible. Aspect ratios can be controlled with the image_config API Parameter","tags":"Image","vendor_id":6,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image"],"output_modalities":["text","image"],"vendor_name":"Google","vendor_icon":"Gemini.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 0.5 + c * 3 + img_o * 60 + ao * 3)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"gemini-2.5-flash-image","created_time":1780786608,"description":"Gemini 2.5 Flash Image, a.k.a. \"Nano Banana,\" is now generally available. It is a state of the art image generation model with contextual understanding. It is capable of image generation, edits, and multi-turn conversations. Aspect ratios can be controlled with the image_config API Parameter","tags":"Image","vendor_id":6,"quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"","completion_ratio":8.333333333333,"image_ratio":100,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image"],"output_modalities":["text","image"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"minimaxmusic-2.6","created_time":1780786443,"tags":"Music","vendor_id":10,"quota_type":1,"model_ratio":0,"model_price":0.2,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["audio"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"minimaxmusic-2.5","created_time":1780786422,"tags":"Music","vendor_id":10,"quota_type":1,"model_ratio":0,"model_price":0.2,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["audio"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"minimaxmusic-2.0","created_time":1780786396,"tags":"Music","vendor_id":10,"quota_type":1,"model_ratio":0,"model_price":0.05,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["audio"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"lyria-3-pro-preview","created_time":1780786348,"description":"Lyria 3 is Google's family of music generation models,you can generate high-quality, 48kHz stereo audio from text prompts or from images. These models deliver structural coherence, including vocals, timed lyrics, and full instrumental arrangements. Lyria 3 Pro can generate full-length songs with verses, choruses, bridges.","tags":"Music","vendor_id":6,"quota_type":1,"model_ratio":0,"model_price":0.1,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["audio"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"lyria-3-clip-preview","created_time":1780786230,"description":"Lyria 3 is Google's family of music generation models, you can generate high-quality, 48kHz stereo audio from text prompts or from images. These models deliver structural coherence, including vocals, timed lyrics, and full instrumental arrangements. Lyria 3 Clip can generate short clips, loops, previews.","tags":"Music","vendor_id":6,"quota_type":1,"model_ratio":0,"model_price":0.05,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["audio"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":322070,"model_name":"qwen3.7-max","created_time":1780159359,"description":"The Max model, which is the largest and most comprehensive in the Qwen3.7 series, currently offers open access to its pure text model capabilities for experience. Qwen3.7 is a new generation of flagship model for the era of intelligent agents, with its core advantage lying in the breadth and depth of intelligent agent capabilities: it excels in various tasks in programming, office and productivity, and long-term autonomous execution.","tags":"Chat","vendor_id":28,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":3,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"Qwen","vendor_icon":"Qwen","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 1.75 + c * 5.15 + cr * 0.2 + cc * 2.15)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":18555,"model_name":"claude-sonnet-4-5","created_time":1779993359,"description":"Claude Sonnet 4.5 is Anthropic’s most advanced Sonnet model to date, optimized for real-world agents and coding workflows. It delivers state-of-the-art performance on coding benchmarks such as SWE-bench Verified, with improvements across system design, code security, and specification adherence. The model is designed for extended autonomous operation, maintaining task continuity across sessions and providing fact-based progress tracking.\n\nSonnet 4.5 also introduces stronger agentic capabilities, including improved tool orchestration, speculative parallel execution, and more efficient context and memory management. With enhanced context tracking and awareness of token usage across tool calls, it is particularly well-suited for multi-context and long-running workflows. Use cases span software engineering, cybersecurity, financial analysis, research agents, and other domains requiring sustained reasoning and tool use.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"p \u003c= 200000 ? tier(\"standard\", p * 3 + c * 15 + cr * 0.3 + cc * 3.75 + cc1h * 6) : tier(\"long_context\", p * 6 + c * 22.5 + cr * 0.6 + cc * 7.5 + cc1h * 12)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1650,"model_name":"claude-opus-4-8","created_time":1779993340,"description":"Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token context window. It is suited for highly autonomous agents, long-horizon agentic work, knowledge work, and memory-driven tasks where coherence over extended sessions matters.\n\nIt is particularly strong on multi-step reasoning, complex coding, and end-to-end project orchestration - large codebases, multi-stage debugging, and long-running asynchronous agent pipelines. Beyond coding, it handles knowledge work such as drafting documents, building presentations, and analyzing data, maintaining quality across very long outputs.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 5 + c * 25 + cr * 0.5 + cc * 6.25 + cc1h * 10)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":3049,"model_name":"claude-opus-4-5","created_time":1779993321,"description":"Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use. It offers strong multimodal capabilities, competitive performance across real-world coding and reasoning benchmarks, and improved robustness to prompt injection. The model is designed to operate efficiently across varied effort levels, enabling developers to trade off speed, depth, and token usage depending on task requirements. It comes with a new parameter to control token efficiency, which can be accessed using the OpenRouter Verbosity parameter with low, medium, or high.\n\nOpus 4.5 supports advanced tool use, extended context management, and coordinated multi-agent setups, making it well-suited for autonomous research, debugging, multi-step planning, and spreadsheet/browser manipulation. It delivers substantial gains in structured reasoning, execution reliability, and alignment compared to prior Opus generations, while reducing token overhead and improving performance on long-running tasks.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 5 + c * 25 + cr * 0.5 + cc * 6.25 + cc1h * 10)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1882,"model_name":"claude-haiku-4-5","created_time":1779993298,"description":"Claude Haiku 4.5 is Anthropic’s fastest and most efficient model, delivering near-frontier intelligence at a fraction of the cost and latency of larger Claude models. Matching Claude Sonnet 4’s performance across reasoning, coding, and computer-use tasks, Haiku 4.5 brings frontier-level capability to real-time and high-volume applications.\n\nIt introduces extended thinking to the Haiku line; enabling controllable reasoning depth, summarized or interleaved thought output, and tool-assisted workflows with full support for coding, bash, web search, and computer-use tools. Scoring \u003e73% on SWE-bench Verified, Haiku 4.5 ranks among the world’s best coding models while maintaining exceptional responsiveness for sub-agents, parallelized execution, and scaled deployment.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["default"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 1 + c * 5 + cr * 0.1 + cc * 1.25 + cc1h * 2)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":6847,"model_name":"claude-opus-4-7","created_time":1779953090,"description":"Opus 4.7 is the next generation of Anthropic's Opus family, built for long-running, asynchronous agents. Building on the coding and agentic strengths of Opus 4.6, it delivers stronger performance on complex, multi-step tasks and more reliable agentic execution across extended workflows. It is especially effective for asynchronous agent pipelines where tasks unfold over time - large codebases, multi-stage debugging, and end-to-end project orchestration.\n\nBeyond coding, Opus 4.7 brings improved knowledge work capabilities - from drafting documents and building presentations to analyzing data. It maintains coherence across very long outputs and extended sessions, making it a strong default for tasks that require persistence, judgment, and follow-through.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 5 + c * 25 + cr * 0.5 + cc * 6.25 + cc1h * 10)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Kling-O1","created_time":1779477952,"description":"Kling Unified Video Model supports multimodal inputs such as text and images, and can understand complex creative instructions through natural language. It is suitable for multimodal reference, subject consistency, complex scene generation, and high consistency video creation.","tags":"Video,Async","vendor_id":4,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Kwai","vendor_icon":"Kling.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":15},{"value":"1080p","multiplier":20},{"value":"2048p","multiplier":30},{"value":"4096p","multiplier":40}]}},{"usage_tokens":6561,"model_name":"doubao-seed-2-0-mini-260428","created_time":1779271112,"description":"Omnimodal understanding model, shorter reasoning length, higher token efficiency.","tags":"Chat","vendor_id":17,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_mode":"tiered_expr","billing_expr":"p \u003c= 128000 ? tier(\"standard\", p * 0.1 + c * 0.4 + cr * 0.02 + cc * 3.75 + ai * 1.5) : tier(\"long_context\", p * 0.2 + c * 0.8 + cr * 0.04 + cc * 7.5 + ai * 3)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":6032,"model_name":"doubao-seed-2-0-lite-260428","created_time":1779271090,"description":"The first omnimodal understanding model in the doubao large model family, supporting native unified understanding of video, images, audio, and text, with upgraded Agent, Coding, and GUI capabilities.","tags":"Chat","vendor_id":17,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_mode":"tiered_expr","billing_expr":"p \u003c= 128000 ? tier(\"standard\", p * 0.25 + c * 2 + cr * 0.05 + cc * 3.75 + ai * 3.75) : tier(\"long_context\", p * 0.5 + c * 4 + cr * 0.1 + cc * 7.5 + ai * 7.5)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"agent-vod-explain-comic","created_time":1779270999,"description":"The AI commentary script-to-video Agent, launched by the Omni team, truly achieves full-process automation. Simply upload a pure text script (script example: ) + episode assets (optional), and generate a finished video with one click (without manual storyboard splitting or asset linking). The availability rate is over 85%;","tags":"Video,Async","vendor_id":23,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["video"],"vendor_name":"Omni","vendor_icon":"https://cos.frostai.cn/omnirouters/logo/logo.png","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":80},{"value":"1080p","multiplier":85}]}},{"usage_tokens":8390766,"model_name":"gemini-3.5-flash","created_time":1779214242,"description":"Gemini 3.5 Flash is Google's high-efficiency multimodal model, bringing near-Pro level coding and reasoning at Flash-tier cost and speed. It is highly optimized for coding proficiency and parallel agentic execution loops, supporting text, image, video, audio, and PDF inputs.\n\nDefaults to medium thinking effort for faster and more cost-efficient responses, with full support for thinking levels (minimal, low, medium, high) for fine-grained cost/performance trade-offs.","tags":"Chat","vendor_id":6,"quota_type":0,"model_ratio":0.75,"model_price":0,"owner_by":"","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":6.66667e-7,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Hunyuan-3d-scene-2.0","created_time":1778430053,"tags":"3D,Async","vendor_id":2,"quota_type":1,"model_ratio":0,"model_price":32,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 3","Tier 4","default","Tier 1","Tier 2"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Tencent","vendor_icon":"Hunyuan.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Hunyuan-3d-panorama-2.0","created_time":1778430034,"tags":"3D","vendor_id":2,"quota_type":1,"model_ratio":0,"model_price":0.8,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 3","Tier 4","default","Tier 1","Tier 2"],"supported_endpoint_types":["openai","openai-video"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Tencent","vendor_icon":"Hunyuan.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Veo-3.1-lite","created_time":1778152919,"tags":"Video,Async","vendor_id":6,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":5.5},{"value":"1080p","multiplier":8.5},{"value":"2048p","multiplier":15},{"value":"4096p","multiplier":18}]}},{"usage_tokens":38698,"model_name":"grok-4.3","created_time":1777601587,"description":"Grok 4.3 is a reasoning model from xAI. It accepts text and image inputs with text output, and is suited for agentic workflows, instruction-following tasks, and applications requiring high factual accuracy. Reasoning is always active and cannot be disabled or configured by effort level.\n\nIt supports a 1 million token context window with no output token limit, making it well-suited for long-document analysis, deep research, and multi-step agentic tasks. Pricing is tiered: requests exceeding 200k total tokens are billed at a higher rate.","tags":"Chat","vendor_id":14,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image"],"output_modalities":["text"],"vendor_name":"xAI","vendor_icon":"https://t0.gstatic.com/faviconV2?client=SOCIAL\u0026type=FAVICON\u0026fallback_opts=TYPE,SIZE,URL\u0026url=https://x.ai/\u0026size=256","billing_mode":"tiered_expr","billing_expr":"p \u003c= 200000 ? tier(\"standard\", p * 1.25 + c * 2.5 + cr * 0.2) : tier(\"long_context\", p * 2.5 + c * 5 + cr * 0.4)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":5455,"model_name":"gpt-oss-120b","created_time":1777536769,"description":"gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized to run on a single H100 GPU with native MXFP4 quantization. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Happyhorse-1.0","created_time":1777525350,"description":"HappyHorse-1.0 supports text-to-video and image-to-video generation, boasts highly realistic dynamic image generation capabilities, can accurately understand text semantics, and produces high-quality videos that are smooth, natural, and rich in detail.","tags":"Video,Async","vendor_id":28,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Qwen","vendor_icon":"Qwen","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":14},{"value":"1080p","multiplier":25},{"value":"2048p","multiplier":32},{"value":"4096p","multiplier":38}]}},{"usage_tokens":915712,"model_name":"gpt-5.5","created_time":1777064148,"description":"GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token context window (922K input, 128K output) with support for text and image inputs, enabling large-scale reasoning, coding, and multimodal workflows within a single system.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","billing_mode":"tiered_expr","billing_expr":"p \u003c= 272000 ? tier(\"standard\", p * 5 + c * 30 + cr * 0.5) : tier(\"long_context\", p * 10 + c * 45 + cr * 1)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":132715,"model_name":"deepseek-v4-flash","created_time":1777008408,"description":"DeepSeek-V4-Flash (Official Release) — Agent capabilities significantly enhanced, with benchmark performance far surpassing V4-Pro-Preview. This model is the cost-effective variant of the V4 series. It also supports a 1M-token ultra-long context window, delivering lower cost and higher response speed while retaining strong reasoning ability. It performs close to the Pro version on simple to moderately complex Agent tasks, making it well suited to latency- and cost-sensitive applications such as dialogue generation, content creation, and lightweight Agent invocation. It supports switching between thinking mode and standard mode, enabling a flexible balance of performance and efficiency.","tags":"Chat","vendor_id":16,"quota_type":0,"model_ratio":0.075,"model_price":0,"owner_by":"","completion_ratio":2,"cache_ratio":2,"enable_groups":["Tier 1","Tier 2","Tier 3","default"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"DeepSeek","vendor_icon":"DeepSeek.Color","billing_mode":"tiered_expr","billing_expr":"(tier(\"base\", p * 0.45 + c * 1.35 + cr * 0.014)) * ((hour(\"Asia/Shanghai\") \u003e= 9 || hour(\"Asia/Shanghai\") \u003c 12) \u0026\u0026 (hour(\"Asia/Shanghai\") \u003e= 14 || hour(\"Asia/Shanghai\") \u003c 18) ? 0.5 : 1)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":177167,"model_name":"deepseek-v4-pro","created_time":1777008376,"description":"deepseek-v4-pro-0813 is DeepSeek’s flagship LLM released on August 13, 2026, featuring a 1M token context window and up to 384K token output. It supports dual-mode inference — thinking mode , making it well-suited for complex agentic tasks, long-document processing, and multi-turn reasoning.","tags":"Chat","vendor_id":16,"quota_type":0,"model_ratio":1.286,"model_price":0,"owner_by":"","completion_ratio":2,"cache_ratio":0.083359253499,"enable_groups":["Tier 3","default","Tier 1","Tier 2"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"DeepSeek","vendor_icon":"DeepSeek.Color","billing_mode":"tiered_expr","billing_expr":"(tier(\"base\", p * 1.35 + c * 4 + cr * 0.45)) * ((hour(\"Asia/Shanghai\") \u003e= 9 || hour(\"Asia/Shanghai\") \u003c 12) \u0026\u0026 (hour(\"Asia/Shanghai\") \u003e= 14 || hour(\"Asia/Shanghai\") \u003c 18) ? 0.5 : 1)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":9994,"model_name":"kimi-k2.6","created_time":1776951232,"description":"Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and can convert prompts and visual inputs into production-ready interfaces. Its agent swarm architecture scales to hundreds of parallel sub-agents for autonomous task decomposition - delivering documents, websites, and spreadsheets in a single run without human oversight.","tags":"Chat","vendor_id":27,"quota_type":0,"model_ratio":0.455,"model_price":0,"owner_by":"","completion_ratio":4.210526315789,"cache_ratio":0.169230769231,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"Kimi","vendor_icon":"Kimi","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":2179,"model_name":"GPT-image-2","created_time":1776938500,"description":"GPT-5.4 Image 2 combines OpenAI's GPT-5.4 model with state-of-the-art image generation capabilities from GPT Image 2. It enables rich multimodal workflows, allowing users to seamlessly move between reasoning, coding, and visual generation within the same interaction.","tags":"Image","vendor_id":5,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","billing_mode":"tiered_expr","billing_expr":"param(\"quality\") == \"low\" \u0026\u0026 param(\"resolution\") == \"1K\"\n  ? tier(\"low-1K\", 250000)\n  : param(\"quality\") == \"low\" \u0026\u0026 param(\"resolution\") == \"2K\"\n  ? tier(\"low-2K\", 285000)\n  : param(\"quality\") == \"low\" \u0026\u0026 param(\"resolution\") == \"4K\"\n  ? tier(\"low-4K\", 350000)\n  : param(\"quality\") == \"medium\" \u0026\u0026 param(\"resolution\") == \"1K\"\n  ? tier(\"medium-1K\", 320000)\n  : param(\"quality\") == \"medium\" \u0026\u0026 param(\"resolution\") == \"2K\"\n  ? tier(\"medium-2K\", 425000)\n  : param(\"quality\") == \"medium\" \u0026\u0026 param(\"resolution\") == \"4K\"\n  ? tier(\"medium-4K\", 500000)\n  : param(\"quality\") == \"high\" \u0026\u0026 param(\"resolution\") == \"1K\"\n  ? tier(\"high-1K\", 550000)\n  : param(\"quality\") == \"high\" \u0026\u0026 param(\"resolution\") == \"2K\"\n  ? tier(\"high-2K\", 800000)\n  : param(\"quality\") == \"high\" \u0026\u0026 param(\"resolution\") == \"4K\"\n  ? tier(\"high-4K\", 1200000)\n  : tier(\"default\", 1200000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Kling-3.0-omni","created_time":1776921453,"description":"Kling 3.0 Omni is an all-in-one multimodal video model of the Ke Ling 3.0 series, which supports input of text, images, videos, etc. It has character voice drive, native audio output, and storyboarding narrative capabilities, suitable for complex storytelling videos, multi subject consistency, and synchronized audio and visual creation.","tags":"Video,Async","vendor_id":4,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image","video"],"output_modalities":["video"],"vendor_name":"Kwai","vendor_icon":"Kling.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":17},{"value":"1080p","multiplier":22},{"value":"2048p","multiplier":30},{"value":"4096p","multiplier":55}]}},{"usage_tokens":1,"model_name":"Kling-3.0","created_time":1776921448,"description":"The latest Keling V3 live video model. Support intelligent storyboarding and 15 second long video generation, with up to 4K Ultra image output, capable of scene switching and continuous narrative, suitable for enterprise advertising and professional film and television creation.","tags":"Video,Async","vendor_id":4,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Kwai","vendor_icon":"Kling.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":17},{"value":"1080p","multiplier":22},{"value":"2048p","multiplier":30},{"value":"4096p","multiplier":55}]}},{"model_name":"Hailuo-2.3-fast","created_time":1776921440,"description":"MiniMax-Hailuo-2.3 is a brand-new video generation model from MiniMax, featuring comprehensive upgrades in body movements, physical representation, and command compliance capabilities. MiniMax-Hailuo-2.3-Fast, as the Fast version of MiniMax-Hailuo-2.3, significantly enhances generation speed, offering a higher cost-performance ratio while maintaining image quality and expressiveness.","tags":"Video,Async","vendor_id":10,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":3.6},{"value":"1080p","multiplier":5.8},{"value":"2048p","multiplier":10},{"value":"4096p","multiplier":14}]}},{"model_name":"Hailuo-2.3","created_time":1776921433,"description":"Minimax Hailuo 2.3 is a newly upgraded video generation model, which excels in body movements, physical effects, and the ability to understand and execute commands.","tags":"Video,Async","vendor_id":10,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 2","Tier 3","Tier 4","default","Tier 1"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":5},{"value":"1080p","multiplier":9},{"value":"2048p","multiplier":14},{"value":"4096p","multiplier":23}]}},{"model_name":"Vidu-q3-turbo","created_time":1776915125,"description":"viduq3-turbo is a multimodal video generation model for film and short-drama creation that supports up to 16 seconds of synchronized audio and video output, enabling multi-shot orchestration and single-take generation with consistent emotional and rhythmic pacing. It features multilingual dialogue and text rendering, complex transitions and camera control capabilities, and can complete narrative expression within a single generation, reducing post-production costs and advancing AI video from fragment generation toward one-click, industrial-scale finished production.","tags":"Video,Async","vendor_id":1,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 2","Tier 3","Tier 4","default","Tier 1"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Vidu","vendor_icon":"Vidu","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":6},{"value":"1080p","multiplier":9},{"value":"2048p","multiplier":12},{"value":"4096p","multiplier":18}],"audio_generation_surcharge":0.08}},{"model_name":"Vidu-q3-pro","created_time":1776915119,"description":"viduq3-pro is a multimodal video generation model for film and short-drama creation that supports up to 16 seconds of synchronized audio and video output, enabling multi-shot orchestration and single-take generation with consistent emotional and rhythmic pacing. It features multilingual dialogue and text rendering, complex transitions and camera control capabilities, and can complete narrative expression within a single generation, reducing post-production costs and advancing AI video from fragment generation toward one-click, industrial-scale finished production.","tags":"Video,Async","vendor_id":1,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 1","Tier 2","Tier 3","Tier 4","default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Vidu","vendor_icon":"Vidu","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":15},{"value":"1080p","multiplier":17},{"value":"2048p","multiplier":22},{"value":"4096p","multiplier":24}],"audio_generation_surcharge":0.08}},{"model_name":"Vidu-q3-mix","created_time":1776915115,"tags":"Video,Async","vendor_id":1,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Vidu","vendor_icon":"Vidu","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":12},{"value":"1080p","multiplier":15},{"value":"2048p","multiplier":18},{"value":"4096p","multiplier":22}],"audio_generation_surcharge":0.08}},{"model_name":"Vidu-q3","created_time":1776915109,"description":"Support reference video capability, can generate videos with consistent subjects based on multiple reference images and prompt words, support intelligent mirror cutting and simultaneous sound and image output, better consistency across multiple camera positions, suitable for multi image reference, subject preservation, and complex scene video creation.","tags":"Video,Async","vendor_id":1,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 4","default","Tier 1","Tier 2","Tier 3"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image","video"],"output_modalities":["video"],"vendor_name":"Vidu","vendor_icon":"Vidu","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":10},{"value":"1080p","multiplier":14},{"value":"2048p","multiplier":22},{"value":"4096p","multiplier":30}],"audio_generation_surcharge":0.08}},{"model_name":"PixVerse-v6","created_time":1776915102,"description":"Support movie grade camera control, native audio generation, and multi lens continuous output, suitable for professional film and television creation and high-quality advertising production.","tags":"Video,Async","vendor_id":30,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"PixVerse","vendor_icon":"PixVerse","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":6},{"value":"1080p","multiplier":12},{"value":"2048p","multiplier":15},{"value":"4096p","multiplier":20}]}},{"model_name":"PixVerse-v5.6","created_time":1776915096,"description":"Further enhanced in terms of cinematic visual aesthetics and vocal expression, with more precise dynamic control, suitable for professional content production with high requirements for visual texture and vocal effects.","tags":"Video,Async","vendor_id":30,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"PixVerse","vendor_icon":"PixVerse","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":5.5},{"value":"1080p","multiplier":11},{"value":"2048p","multiplier":15},{"value":"4096p","multiplier":20}]}},{"model_name":"PixVerse-c1","created_time":1776915090,"description":"Deeply optimized for the film and television industry, it supports multimodal inputs such as text, images, start and end frames, and multi grid storyboarding references. It can produce 15 second 1080P synchronized audio and visual videos, with industrial grade action performance and film and television grade special effects rendering capabilities. It is suitable for short dramas, anime, fantasy special effects, and pre film storyboarding production scenes.","tags":"Video,Async","vendor_id":30,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"PixVerse","vendor_icon":"PixVerse","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":6.5},{"value":"1080p","multiplier":14},{"value":"2048p","multiplier":18},{"value":"4096p","multiplier":20}]}},{"model_name":"Hunyuan-1.5","created_time":1776915064,"description":"Supporting multimodal input of text and images to generate high-definition videos, it can achieve scene switching and multi role interaction, simplify the production process, reduce costs, and be applied in enterprise advertising marketing and personal creative landing scenarios.","tags":"Video,Async","vendor_id":2,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 4","default","Tier 1","Tier 2","Tier 3"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Tencent","vendor_icon":"Hunyuan.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":4.5},{"value":"1080p","multiplier":7.5},{"value":"2048p","multiplier":12},{"value":"4096p","multiplier":17}]}},{"model_name":"Vidu:q2","created_time":1776912705,"description":"Support for reference images, text images, and image editing, precise rendering of Chinese and English text, pixel level restoration of UI/chart design details, suitable for creating posters, infographics, and more.","tags":"Image","vendor_id":1,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai-video","openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Vidu","vendor_icon":"Vidu","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 46000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 95000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 140000)\n  : tier(\"default\", 140000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Seedream:5.0-lite","created_time":1776912698,"tags":"Image","vendor_id":17,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["Tier 1","default"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 38000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 38000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 38000)\n  : tier(\"default\", 38000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Seedream:4.5","created_time":1776912688,"tags":"Image","vendor_id":17,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["Tier 1","default"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 40000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 40000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 40000)\n  : tier(\"default\", 40000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Seedream:4.0","created_time":1776912678,"tags":"Image","vendor_id":17,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 38000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 38000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 38000)\n  : tier(\"default\", 38000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":5,"model_name":"Nano-Banana2","created_time":1776912661,"tags":"Image","vendor_id":6,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Google","vendor_icon":"Gemini.Color","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 75000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 150000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 165000)\n  : tier(\"default\", 165000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Nano-Banana-Pro","created_time":1776912653,"tags":"Image","vendor_id":6,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Google","vendor_icon":"Gemini.Color","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 150000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 150000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 270000)\n  : tier(\"default\", 270000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Hunyuan:3.0","created_time":1776912623,"description":"Based on the Hybrid Large Model, it can consider the layout, composition, and brushstrokes of images, and use world knowledge to infer common sense images. At the same time, it can analyze complex semantics at the thousand-word level, generate long texts, complex comics, and emojis, and also produce vivid and interesting science illustrations.\nModel price","tags":"Image","vendor_id":2,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Tencent","vendor_icon":"Hunyuan.Color","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 30000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 45000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 55000)\n  : tier(\"default\", 550000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Kling:O1","created_time":1776908376,"tags":"Image","vendor_id":4,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1","Tier 2"],"supported_endpoint_types":["openai-video","openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Kwai","vendor_icon":"Kling.Color","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 30000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 30000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 60000)\n  : tier(\"default\", 60000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Kling:3.0-omni","created_time":1776908362,"tags":"Image","vendor_id":4,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1","Tier 2"],"supported_endpoint_types":["openai-video","openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Kwai","vendor_icon":"Kling.Color","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 30000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 30000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 60000)\n  : tier(\"default\", 60000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":3,"model_name":"Kling:3.0","created_time":1776840090,"tags":"Image","vendor_id":4,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["Tier 2","default","Tier 1"],"supported_endpoint_types":["openai-video","openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Kwai","vendor_icon":"Kling.Color","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 30000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 30000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 60000)\n  : tier(\"default\", 60000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"Veo-3.1-fast","created_time":1776829721,"tags":"Video,Async","vendor_id":6,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":18},{"value":"1080p","multiplier":27},{"value":"2048p","multiplier":40},{"value":"4096p","multiplier":60}]}},{"model_name":"Veo-3.1","created_time":1776829702,"tags":"Video,Async","vendor_id":6,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Google","vendor_icon":"Gemini.Color","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"720p","multiplier":45},{"value":"1080p","multiplier":70},{"value":"2048p","multiplier":105},{"value":"4096p","multiplier":160}]}},{"usage_tokens":2005,"model_name":"gpt-5.1","created_time":1775750707,"description":"GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5. It uses adaptive reasoning to allocate computation dynamically, responding quickly to simple queries while spending more depth on complex tasks. The model produces clearer, more grounded explanations with reduced jargon, making it easier to follow even on technical or multi-step problems.\n\nBuilt for broad task coverage, GPT-5.1 delivers consistent gains across math, coding, and structured analysis workloads, with more coherent long-form answers and improved tool-use reliability. It also features refined conversational alignment, enabling warmer, more intuitive responses without compromising precision. GPT-5.1 serves as the primary full-capability successor to GPT-5","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":2093,"model_name":"gpt-5","created_time":1775750692,"description":"GPT-5 is OpenAI’s most advanced model, offering major improvements in reasoning, code quality, and user experience. It is optimized for complex tasks that require step-by-step reasoning, instruction following, and accuracy in high-stakes use cases. It supports test-time routing features and advanced prompt understanding, including user-specified intent like \"think hard about this.\" Improvements include reductions in hallucination, sycophancy, and better performance in coding, writing, and health-related tasks.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":0.625,"model_price":0,"owner_by":"","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":6363,"model_name":"gpt-4o","created_time":1775750661,"description":"The 2024-11-20 version of GPT-4o offers a leveled-up creative writing ability with more natural, engaging, and tailored writing to improve relevance \u0026 readability. It’s also better at working with uploaded files, providing deeper insights \u0026 more thorough responses.\n\nGPT-4o (\"o\" for \"omni\") is OpenAI's latest AI model, supporting both text and image inputs with text outputs. It maintains the intelligence level of GPT-4 Turbo while being twice as fast and 50% more cost-effective. GPT-4o also offers improved performance in processing non-English languages and enhanced visual capabilities.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"","completion_ratio":4,"cache_ratio":0.5,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":4679,"model_name":"gpt-5-nano","created_time":1775750617,"description":"GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments. While limited in reasoning depth compared to its larger counterparts, it retains key instruction-following and safety features. It is the successor to GPT-4.1-nano and offers a lightweight option for cost-sensitive or real-time applications.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":0.025,"model_price":0,"owner_by":"","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1900,"model_name":"glm-5.1","created_time":1775697180,"description":"GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks. Unlike previous models built around minute-level interactions, GLM-5.1 can work independently and continuously on a single task for more than 8 hours, autonomously planning, executing, and improving itself throughout the process, ultimately delivering complete, engineering-grade results.","tags":"Chat","vendor_id":18,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":3.142857142857,"cache_ratio":0.185714285714,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"Z.ai","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/acc3a4676ad771bd616079c832b3d7a5_1779342707107.svg","billing_mode":"tiered_expr","billing_expr":"len \u003c= 32000 ? tier(\"standard\", p * 0.84 + c * 3.36 + cr * 0.182) : tier(\"long_context\", p * 1.12 + c * 3.92 + cr * 0.28)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":332146,"model_name":"minimax-m2.7","created_time":1775610018,"description":"MiniMax-M2.7 is a next-generation large language model designed for autonomous, real-world productivity and continuous improvement. Built to actively participate in its own evolution, M2.7 integrates advanced agentic capabilities through multi-agent collaboration, enabling it to plan, execute, and refine complex tasks across dynamic environments.\n\nTrained for production-grade performance, M2.7 handles workflows such as live debugging, root cause analysis, financial modeling, and full document generation across Word, Excel, and PowerPoint. It delivers strong results on benchmarks including 56.2% on SWE-Pro and 57.0% on Terminal Bench 2, while achieving a 1495 ELO on GDPval-AA, setting a new standard for multi-agent systems operating in real-world digital workflows.","tags":"Chat","vendor_id":10,"quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"","completion_ratio":4,"cache_ratio":0.2,"create_cache_ratio":1.333333333333,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":58302,"model_name":"qwen3.6-plus","created_time":1775608675,"tags":"Chat","vendor_id":28,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image"],"output_modalities":["text"],"vendor_name":"Qwen","vendor_icon":"Qwen","billing_mode":"tiered_expr","billing_expr":"len \u003c= 256000 ? tier(\"standard\", p * 0.3 + c * 1.72 + cr * 0.03 + cc * 0.35) : tier(\"long_context\", p * 1.15 + c * 6.85 + cr * 0.12 + cc * 1.45)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"doubao-seedance-2-0-fast-260128","created_time":1775111830,"description":"Seedance 2.0 Fast is a video generation model from ByteDance. It supports text-to-video, image-to-video with first and last frame control, and multimodal reference-to-video. It prioritizes generation speed and lower cost over maximum output quality.","tags":"Video,Async","vendor_id":17,"quota_type":0,"model_ratio":2.8,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai-video","openai"],"input_modalities":["text","image","audio","video"],"output_modalities":["video"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_usage_schema":{"resolution":{"enum":["480p","720p"],"description":{"en":"Output video resolution","zh":"输出视频分辨率"},"enumLabels":{"480p":{"en":"480p","zh":"480p"},"720p":{"en":"720p","zh":"720p"}}},"tokens":{"type":"number","unit":"token","description":{"en":"Billing token unit price","zh":"计费 Token 单价"}},"video_input":{"enum":["none","video"],"description":{"en":"Reference video input","zh":"参考视频输入"},"enumLabels":{"none":{"en":"No reference video","zh":"无参考视频"},"video":{"en":"With reference video","zh":"有参考视频"}}}},"billing_usage_examples":[{"label":"480p · 5s","facts":{"resolution":"480p","tokens":48038,"video_input":"none"}},{"label":"720p · 5s","facts":{"resolution":"720p","tokens":108000,"video_input":"none"}},{"label":"720p · 5s (+4s 输入视频)","facts":{"resolution":"720p","tokens":194400,"video_input":"video"}}],"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"doubao-seedance-2-0-260128","created_time":1775111811,"description":"Seedance 2.0 is a video generation model from ByteDance. It supports text-to-video, image-to-video with first and last frame control, and multimodal reference-to-video. It is particularly strong at preserving character consistency, visual style, and camera movement from reference material.","tags":"Video,Async","vendor_id":17,"quota_type":0,"model_ratio":3.75,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai-video","openai"],"input_modalities":["text","image","audio","video"],"output_modalities":["video"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_usage_schema":{"resolution":{"enum":["480p","720p","1080p","4k"],"description":{"en":"Output video resolution","zh":"输出视频分辨率"},"enumLabels":{"1080p":{"en":"1080p","zh":"1080p"},"480p":{"en":"480p","zh":"480p"},"4k":{"en":"4k","zh":"4k"},"720p":{"en":"720p","zh":"720p"}}},"tokens":{"type":"number","unit":"token","description":{"en":"Billing token unit price","zh":"计费 Token 单价"}},"video_input":{"enum":["none","video"],"description":{"en":"Reference video input","zh":"参考视频输入"},"enumLabels":{"none":{"en":"No reference video","zh":"无参考视频"},"video":{"en":"With reference video","zh":"有参考视频"}}}},"billing_usage_examples":[{"label":"480p · 5s","facts":{"resolution":"480p","tokens":48038,"video_input":"none"}},{"label":"720p · 5s","facts":{"resolution":"720p","tokens":108000,"video_input":"none"}},{"label":"1080p · 5s","facts":{"resolution":"1080p","tokens":243000,"video_input":"none"}},{"label":"4k · 5s","facts":{"resolution":"4k","tokens":972000,"video_input":"none"}},{"label":"720p · 5s (+4s 输入视频)","facts":{"resolution":"720p","tokens":194400,"video_input":"video"}}],"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":68,"model_name":"vidu-tts-1.0","created_time":1774802131,"tags":"tts","vendor_id":1,"quota_type":0,"model_ratio":45,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai-video","openai","openai-tts"],"input_modalities":["text"],"output_modalities":["speech"],"vendor_name":"Vidu","vendor_icon":"Vidu","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"speech-2.8-turbo","created_time":1774793568,"description":"The MiniMax Speech 2.8 Turbo is optimized based on the 2.8 HD same generation architecture, significantly reducing inference latency while maintaining high sound quality. It supports emotional tone labels and 40+languages, with low latency and high concurrency, making it suitable for large-scale concurrent scenarios such as real-time conversations and live interactions that are sensitive to response speed.","tags":"tts","vendor_id":10,"quota_type":0,"model_ratio":21.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai","openai-tts"],"input_modalities":["text"],"output_modalities":["speech"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":159792,"model_name":"speech-2.8-hd","created_time":1774793532,"description":"MiniMax Speech 2.8 HD is the latest generation of high fidelity speech synthesis model from MiniMax, which has been fully upgraded in terms of sound quality delicacy, emotional expression, and multilingual coverage. Supporting emotional tone label control and 40+languages, the Chinese expression is more natural and the rhythm is closer to real people, suitable for professional scenarios with high requirements for sound quality and expressiveness, such as audiobook recording, intelligent customer service, short video dubbing, etc.","tags":"tts","vendor_id":10,"quota_type":0,"model_ratio":29,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai","openai-tts"],"input_modalities":["text"],"output_modalities":["speech"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"speech-02-turbo","created_time":1774793522,"description":"The MiniMax Speech 02 Turbo further compresses delay while preserving natural rhythm, balancing high tone similarity and low delay, making it suitable for real-time speech applications that require response speed but still have a certain level of tone quality.","tags":"tts","vendor_id":10,"quota_type":0,"model_ratio":21.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai","openai-tts"],"input_modalities":["text"],"output_modalities":["speech"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"speech-02-hd","created_time":1774793514,"description":"The MiniMax Speech 02 HD features high tone similarity and natural rhythm expression, supports multilingual synthesis, and performs robustly in long text coherence and pronunciation stability, making it suitable for high-quality dubbing, audiobook production, and other content production scenarios.","tags":"tts","vendor_id":10,"quota_type":0,"model_ratio":29,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai","openai-tts"],"input_modalities":["text"],"output_modalities":["audio"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":3,"model_name":"Nano-Banana","created_time":1774592115,"description":"Gemini 2.5 Flash Image, a.k.a. \"Nano Banana,\" is now generally available. It is a state of the art image generation model with contextual understanding. It is capable of image generation, edits, and multi-turn conversations.","tags":"Image","vendor_id":6,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Google","vendor_icon":"Gemini.Color","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 45000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 55000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 65000)\n  : tier(\"default\", 65000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"speech-2.6-turbo","created_time":1774473589,"description":"MiniMax Speech 2.6 Turbo is optimized for real-time interactive scenarios, with end-to-end latency of less than 250ms, maximizing throughput while ensuring sound quality availability. It is suitable for scenarios that require fast response, such as real-time voice conversations and online educational interactions.","tags":"tts","vendor_id":10,"quota_type":0,"model_ratio":21.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai","openai-tts"],"input_modalities":["text"],"output_modalities":["speech"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"speech-2.6-hd","created_time":1774473582,"description":"MiniMax Speech 2.6 HD is a mature and stable high fidelity speech synthesis model with ultra-low latency output, clear and natural sound quality, support for multiple languages and rich timbre libraries, suitable for scenarios requiring high-quality speech output such as Voice Agent intelligent agent interaction and professional content production.","tags":"tts","vendor_id":10,"quota_type":0,"model_ratio":21.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai","openai-tts"],"input_modalities":["text"],"output_modalities":["speech"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"seed-tts-2.0","created_time":1774473572,"tags":"tts","vendor_id":17,"quota_type":0,"model_ratio":36,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai","openai-tts"],"input_modalities":["text"],"output_modalities":["speech"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":134,"model_name":"seed-tts-1.0","created_time":1774473559,"tags":"tts","vendor_id":17,"quota_type":0,"model_ratio":36,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai","openai-tts"],"input_modalities":["text"],"output_modalities":["speech"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":60,"model_name":"MJ:v7","created_time":1774191871,"tags":"Image","vendor_id":15,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai","image-generation"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"Midjourney","vendor_icon":"Midjourney","billing_mode":"tiered_expr","billing_expr":"param(\"resolution\") == \"1K\"\n  ? tier(\"1K\", 68000)\n  : param(\"resolution\") == \"2K\"\n  ? tier(\"2K\", 75000)\n  : param(\"resolution\") == \"4K\"\n  ? tier(\"4K\", 80000)\n  : tier(\"default\", 80000)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":491084,"model_name":"gpt-5.4-nano","created_time":1773790619,"description":"GPT-5.4 nano is the most lightweight and cost-efficient variant of the GPT-5.4 family, optimized for speed-critical and high-volume tasks. It supports text and image inputs and is designed for low-latency use cases such as classification, data extraction, ranking, and sub-agent execution.\n\nThe model prioritizes responsiveness and efficiency over deep reasoning, making it ideal for pipelines that require fast, reliable outputs at scale. GPT-5.4 nano is well suited for background tasks, real-time systems, and distributed agent architectures where minimizing cost and latency is essential.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"","completion_ratio":6.25,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":465232,"model_name":"gpt-5.4-mini","created_time":1773790588,"description":"GPT-5.4 mini brings the core capabilities of GPT-5.4 to a faster, more efficient model optimized for high-throughput workloads. It supports text and image inputs with strong performance across reasoning, coding, and tool use, while reducing latency and cost for large-scale deployments.\n\nThe model is designed for production environments that require a balance of capability and efficiency, making it well suited for chat applications, coding assistants, and agent workflows that operate at scale. GPT-5.4 mini delivers reliable instruction following, solid multi-step reasoning, and consistent performance across diverse tasks with improved cost efficiency.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":0.375,"model_price":0,"owner_by":"","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"agent-vod-replicate","created_time":1773447843,"description":"The video replication agent developed by the Omni team, after video replication, you will get: One-click access to popular videos - the model automatically understands key elements based on the input video and generates videos similar to the original one  Avoid homogenized content - the model automatically understands the original video style, actions, camera movements, etc., and generates videos that are highly similar to the original but not exactly the same. You can also add optimization directions after replication through prompt words  BGM free choice - you can choose to retain the original video music or remove it, and the model can accurately understand both options","tags":"Video,Async","vendor_id":23,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["video"],"vendor_name":"Omni","vendor_icon":"https://cos.frostai.cn/omnirouters/logo/logo.png","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"540p","multiplier":10},{"value":"720p","multiplier":12},{"value":"1080p","multiplier":14}]}},{"model_name":"agent-vod-mv","created_time":1773447836,"description":"The video music agent developed by the Omni team offers low-threshold and efficient generation: simply upload one non-pure music audio + 1-7 reference images, match with lyrics/prompt words, and without requiring professional editing/modeling skills, you can produce a complete narrative MV in one step with just one click, significantly reducing the creative threshold and time cost.\n🔸Lyrics-driven precise storytelling: Automatically dissect lyrics' plots and emotional keywords, weave them into a coherent storyline, and ensure that the visuals and imagery are highly aligned with the lyrics' artistic conception, avoiding fragmentation. This allows the MV to possess both \"audio-visual appeal\" and \"storytelling quality\", eliminating the need for additional plot construction.\n🔸High fidelity adaptation based on reference images: Accurately reproduce the appearance, clothing, temperament, as well as style, tone, filters, and scene elements of the characters in the reference images. Support unified style across multiple images, eliminate image fragmentation, and ensure that the generated effect is consistent with expectations.\n🔸Seamless integration of audio and visuals: Automatically achieving precise synchronization of lyrics and subtitles, with camera cuts/action transitions fitting the beats and melodic fluctuations of the music, while also supporting subtitles.","tags":"Video,Async","vendor_id":23,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["video"],"vendor_name":"Omni","vendor_icon":"https://cos.frostai.cn/omnirouters/logo/logo.png","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"540p","multiplier":4},{"value":"720p","multiplier":6},{"value":"1080p","multiplier":8}]}},{"usage_tokens":32977,"model_name":"minimax-m2.5","created_time":1773344458,"description":"MiniMax-M2.5 is the latest large language model launched by MiniMax, which has undergone extensive reinforcement learning training in hundreds of thousands of complex real-world environments. The model adopts a MoE architecture and boasts 229 billion parameters. It achieves industry-leading performance in tasks such as programming, intelligent agent tool invocation and search, and office scenarios. It scored 80.2% on SWE-Bench Verified and has an inference speed 37% faster than its predecessor, M2.1","tags":"Chat","vendor_id":10,"quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"","completion_ratio":4,"cache_ratio":0.1,"create_cache_ratio":1.333333333333,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"MiniMax","vendor_icon":"Minimax","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1635,"model_name":"glm-5","created_time":1773344275,"description":"GLM-5 is a new generation of large language model launched by ZhiPu, focusing on complex systems engineering and long-cycle Agent tasks. The model scale has expanded from 355B parameters (32B activations) of GLM-4.5 to 744B parameters (40B activations), and the pre-training data has increased from 23T to 28.5T tokens. GLM-5 integrates DeepSeek Sparse Attention (DSA), significantly reducing deployment costs while maintaining long-context capabilities. In terms of reasoning, programming, and Agent tasks, GLM-5 achieves the best level among open-source models in its category","tags":"Chat","vendor_id":18,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":3.315789473684,"cache_ratio":0.5,"enable_groups":["default","Tier 1"],"supported_endpoint_types":["openai"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"Z.ai","vendor_icon":"https://cdnv2-cache.udelivrs.com/2026/05/acc3a4676ad771bd616079c832b3d7a5_1779342707107.svg","billing_mode":"tiered_expr","billing_expr":"len \u003c= 32000 ? tier(\"standard\", p * 0.56 + c * 2.52 + cr * 0.14) : tier(\"long_context\", p * 0.84 + c * 3.08 + cr * 0.21)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"agent-vod-general","created_time":1773175693,"description":"The universal Agent video generation model developed by the Omni team supports up to 7 reference subject inputs, with an output resolution of 1080p, synchronized audio and video, multi-aspect ratio adaptation, and the ability to generate videos up to 180 seconds in length per generation.","tags":"Video,Async","vendor_id":23,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default","Tier 1","Tier 2","Tier 3","Tier 4"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image"],"output_modalities":["video"],"vendor_name":"Omni","vendor_icon":"https://cos.frostai.cn/omnirouters/logo/logo.png","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"1080p","multiplier":30}]}},{"model_name":"agent-vod-ecommerce","created_time":1773175674,"description":"The e-commerce Agent video generation model developed by the Omni team supports up to 7 reference subject inputs, with an output resolution of 1080p, synchronized audio and video, multi-aspect ratio adaptation, and the ability to generate videos up to 60 seconds in length per generation.","tags":"Video,Async","vendor_id":23,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["Tier 1","Tier 2","Tier 3","Tier 4","default"],"supported_endpoint_types":["openai-video"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["video"],"vendor_name":"Omni","vendor_icon":"https://cos.frostai.cn/omnirouters/logo/logo.png","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1},"sora_per_request_pricing":{"enabled":true,"resolution_tiers":[{"value":"1080p","multiplier":30}]}},{"usage_tokens":27674,"model_name":"gpt-5.4","created_time":1772761047,"description":"GPT-5.4 is OpenAI’s latest frontier model, unifying the Codex and GPT lines into a single system. It features a 1M+ token context window (922K input, 128K output) with support for text and image inputs, enabling high-context reasoning, coding, and multimodal analysis within the same workflow.\n\nThe model delivers improved performance in coding, document understanding, tool use, and instruction following. It is designed as a strong default for both general-purpose tasks and software engineering, capable of generating production-quality code, synthesizing information across multiple sources, and executing complex multi-step workflows with fewer iterations and greater token efficiency.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":6,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","billing_mode":"tiered_expr","billing_expr":"p \u003c= 272000 ? tier(\"standard\", p * 2.5 + c * 15 + cr * 0.25) : tier(\"long_context\", p * 5 + c * 22.5 + cr * 0.5)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":4,"model_name":"text-embedding-ada-002","created_time":1771868305,"description":"text-embedding-ada-002 is an old version of OpenAI's text embedding model.","tags":"Embeddings","vendor_id":5,"quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["embeddings","openai"],"input_modalities":["text"],"output_modalities":["embedding"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":2,"model_name":"text-embedding-3-large","created_time":1771868233,"description":"text-embedding-3-large is OpenAI's most powerful embedding model, suitable for both English and non-English tasks. Embedding is a numerical representation of text that can be used to measure the relevance between two pieces of text. Embeddings are useful in search, clustering, recommendation, anomaly detection, and classification tasks.","tags":"Embeddings","vendor_id":5,"quota_type":0,"model_ratio":0.065,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["embeddings","openai"],"input_modalities":["text"],"output_modalities":["embedding"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"suno_music","created_time":1771867413,"description":"Suno API, this model is used for generating songs and supports custom mode, inspiration mode, etc","tags":"Music,Async","vendor_id":20,"quota_type":1,"model_ratio":0,"model_price":0.2,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai","suno"],"input_modalities":["text"],"output_modalities":["audio"],"vendor_name":"Suno","vendor_icon":"Suno","billing_usage_schema":{"action":{"enum":["music","lyrics"],"description":{"en":"Suno generation action.","zh":"Suno 生成动作。"}},"clips":{"type":"number","unit":"count","description":{"en":"Number of generated music or lyrics clips.","zh":"生成的音乐或歌词片段数量。"}}},"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"suno_lyrics","created_time":1771867400,"description":"Suno API, this model is used for generating lyrics","tags":"Music,Async","vendor_id":20,"quota_type":1,"model_ratio":0,"model_price":0.01,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["openai","suno"],"input_modalities":["text"],"output_modalities":["text"],"vendor_name":"Suno","vendor_icon":"Suno","billing_usage_schema":{"action":{"enum":["music","lyrics"],"description":{"en":"Suno generation action.","zh":"Suno 生成动作。"}},"clips":{"type":"number","unit":"count","description":{"en":"Number of generated music or lyrics clips.","zh":"生成的音乐或歌词片段数量。"}}},"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"doubao-seedream-4-5-251128","created_time":1771867352,"description":"Seedream 4.5 is the latest image multimodal model launched by ByteDance, integrating capabilities such as text-to-image, image-to-image, and group image output, and fusing common sense and reasoning abilities. Compared to the previous generation 4.0 model, its generation effect has been significantly improved, featuring better editorial consistency and multi-image fusion effects. It can more precisely control image details, generate smaller text and smaller human faces more naturally, achieve more harmonious image layout and color, and enhance aesthetic perception","tags":"Image","vendor_id":17,"quota_type":1,"model_ratio":0,"model_price":0.04,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_usage_schema":{"images_above_1_5k":{"type":"number","unit":"count","unitLabel":{"en":"image","fr":"image","ja":"枚","ru":"изображение","vi":"ảnh","zh":"张","zh-TW":"張"},"description":{"en":"Image generation unit price (above 1.5K)","zh":"图片生成单价（1.5K 以上）"}},"images_up_to_1_5k":{"type":"number","unit":"count","unitLabel":{"en":"image","fr":"image","ja":"枚","ru":"изображение","vi":"ảnh","zh":"张","zh-TW":"張"},"description":{"en":"Image generation unit price (1.5K and below)","zh":"图片生成单价（1.5K 及以下）"}},"input_images":{"type":"number","unit":"count","unitLabel":{"en":"image","fr":"image","ja":"枚","ru":"изображение","vi":"ảnh","zh":"张","zh-TW":"張"},"description":{"en":"Input image unit price","zh":"输入图片单价"}},"layer_decomposition":{"type":"boolean","description":{"en":"Whether layer decomposition is enabled","zh":"是否开启图层拆分"}}},"billing_usage_examples":[{"label":"2K · 1 张","facts":{"images_above_1_5k":1,"images_up_to_1_5k":0,"input_images":0,"layer_decomposition":false}},{"label":"1K · 1 张 · 2 张参考图","facts":{"images_above_1_5k":0,"images_up_to_1_5k":1,"input_images":2,"layer_decomposition":false}},{"label":"2K · 4 张组图","facts":{"images_above_1_5k":4,"images_up_to_1_5k":0,"input_images":0,"layer_decomposition":false}},{"label":"图层拆分 · 2K 底图 + 4 层 1.5K","facts":{"images_above_1_5k":1,"images_up_to_1_5k":4,"input_images":1,"layer_decomposition":true}}],"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"model_name":"doubao-seedream-4-0-250828","created_time":1771867328,"description":"Seedream 4.0 is a state-of-the-art multimodal image creation model based on a leading architecture. Its ability to generate aesthetic sense, follow instructions, maintain structural integrity, and ensure subject consistency is at the forefront of the world. The model adopts the same architecture to unify text-to-image generation and editing capabilities, natively supports text, single image, and multiple image inputs, and can automatically adapt to the optimal image ratio and generation quantity through deep reasoning of prompts. It can output up to 15 content-related images in a single continuous output, supporting 4K ultra-high-definition output","tags":"Image","vendor_id":17,"quota_type":1,"model_ratio":0,"model_price":0.03,"owner_by":"","completion_ratio":0,"enable_groups":["default"],"supported_endpoint_types":["image-generation","openai"],"input_modalities":["text","image"],"output_modalities":["image"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_usage_schema":{"images_above_1_5k":{"type":"number","unit":"count","unitLabel":{"en":"image","fr":"image","ja":"枚","ru":"изображение","vi":"ảnh","zh":"张","zh-TW":"張"},"description":{"en":"Image generation unit price (above 1.5K)","zh":"图片生成单价（1.5K 以上）"}},"images_up_to_1_5k":{"type":"number","unit":"count","unitLabel":{"en":"image","fr":"image","ja":"枚","ru":"изображение","vi":"ảnh","zh":"张","zh-TW":"張"},"description":{"en":"Image generation unit price (1.5K and below)","zh":"图片生成单价（1.5K 及以下）"}},"input_images":{"type":"number","unit":"count","unitLabel":{"en":"image","fr":"image","ja":"枚","ru":"изображение","vi":"ảnh","zh":"张","zh-TW":"張"},"description":{"en":"Input image unit price","zh":"输入图片单价"}},"layer_decomposition":{"type":"boolean","description":{"en":"Whether layer decomposition is enabled","zh":"是否开启图层拆分"}}},"billing_usage_examples":[{"label":"2K · 1 张","facts":{"images_above_1_5k":1,"images_up_to_1_5k":0,"input_images":0,"layer_decomposition":false}},{"label":"1K · 1 张 · 2 张参考图","facts":{"images_above_1_5k":0,"images_up_to_1_5k":1,"input_images":2,"layer_decomposition":false}},{"label":"2K · 4 张组图","facts":{"images_above_1_5k":4,"images_up_to_1_5k":0,"input_images":0,"layer_decomposition":false}},{"label":"图层拆分 · 2K 底图 + 4 层 1.5K","facts":{"images_above_1_5k":1,"images_up_to_1_5k":4,"input_images":1,"layer_decomposition":true}}],"model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":6584,"model_name":"doubao-seed-2-0-code-preview-260215","created_time":1771866173,"description":"The Coding model, optimized for real-world programming environments, can stably utilize tools in common IDEs such as Claude Code. The model has been specially optimized for front-end capabilities, ensuring excellent performance when using common front-end frameworks. The model supports the use of Skills and can be utilized in conjunction with various custom skills.","tags":"Chat","vendor_id":17,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_mode":"tiered_expr","billing_expr":"p \u003c= 128000 ? tier(\"standard\", p * 0.5 + c * 3 + cr * 0.1) : tier(\"long_context\", p * 1 + c * 6 + cr * 0.2)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":9683,"model_name":"doubao-seed-2-0-pro-260215","created_time":1771866014,"description":"A flagship-level, all-around general model designed for complex reasoning and long-chain task execution scenarios in the era of Agents. It emphasizes multimodal understanding, long-context reasoning, structured generation, and tool-enhanced execution. With outstanding capabilities in handling complex instructions and executing under multiple constraints, it can stably cope with scenarios such as multi-step complex planning, complex text-image reasoning, video content understanding, and challenging analysis.","tags":"Chat","vendor_id":17,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","video"],"output_modalities":["text"],"vendor_name":"ByteDance","vendor_icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg","billing_mode":"tiered_expr","billing_expr":"p \u003c= 128000 ? tier(\"standard\", p * 0.5 + c * 3 + cr * 0.1) : tier(\"long_context\", p * 1 + c * 6 + cr * 0.2)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":2447,"model_name":"claude-opus-4-6","created_time":1771865013,"description":"Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks. It is built for agents that operate across entire workflows rather than single prompts, making it especially effective for large codebases, complex refactors, and multi-step debugging that unfolds over time. The model shows deeper contextual understanding, stronger problem decomposition, and greater reliability on hard engineering tasks than prior generations.\n\nBeyond coding, Opus 4.6 excels at sustained knowledge work. It produces near-production-ready documents, plans, and analyses in a single pass, and maintains coherence across very long outputs and extended sessions. This makes it a strong default for tasks that require persistence, judgment, and follow-through, such as technical design, migration planning, and end-to-end project execution.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 5 + c * 25 + cr * 0.5 + cc * 6.25 + cc1h * 10)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":2121,"model_name":"claude-sonnet-4-6","created_time":1771864873,"description":"Sonnet 4.6 is Anthropic's most capable Sonnet-class model yet, with frontier performance across coding, agents, and professional work. It excels at iterative development, complex codebase navigation, end-to-end project management with memory, polished document creation, and confident computer use for web QA and workflow automation.","tags":"Chat","vendor_id":13,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["default","Economy Mode"],"supported_endpoint_types":["anthropic","openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"Anthropic","vendor_icon":"Claude.Color","billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 3 + c * 15 + cr * 0.3 + cc * 3.75 + cc1h * 6)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1995,"model_name":"gpt-5.2","created_time":1771780480,"description":"GPT-5.2 is the latest frontier-grade model in the GPT-5 series, offering stronger agentic and long context perfomance compared to GPT-5.1. It uses adaptive reasoning to allocate computation dynamically, responding quickly to simple queries while spending more depth on complex tasks.\n\nBuilt for broad task coverage, GPT-5.2 delivers consistent gains across math, coding, sciende, and tool calling workloads, with more coherent long-form answers and improved tool-use reliability.","tags":"Chat","vendor_id":5,"quota_type":0,"model_ratio":0.875,"model_price":0,"owner_by":"","completion_ratio":8,"cache_ratio":0.1,"enable_groups":["default"],"supported_endpoint_types":["openai"],"input_modalities":["text","image","file"],"output_modalities":["text"],"vendor_name":"OpenAI","vendor_icon":"OpenAI","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}},{"usage_tokens":1142698,"model_name":"gemini-3.1-pro-preview","created_time":1771779925,"description":"Gemini 3.1 Pro Preview is Google’s frontier reasoning model, delivering enhanced software engineering performance, improved agentic reliability, and more efficient token usage across complex workflows. Building on the multimodal foundation of the Gemini 3 series, it combines high-precision reasoning across text, image, video, audio, and code with a 1M-token context window. The 3.1 update introduces measurable gains in SWE benchmarks and real-world coding environments, along with stronger autonomous task execution in structured domains such as finance and spreadsheet-based workflows.\n\nDesigned for advanced development and agentic systems, Gemini 3.1 Pro Preview improves long-horizon stability and tool orchestration while increasing token efficiency. It introduces a new medium thinking level to better balance cost, speed, and performance. The model excels in agentic coding, structured planning, multimodal analysis, and workflow automation, making it well-suited for autonomous agents, financial modeling, spreadsheet automation, and high-context enterprise tasks.","tags":"Chat","vendor_id":6,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":4,"enable_groups":["default"],"supported_endpoint_types":["gemini","openai"],"input_modalities":["text","image","file","audio","video"],"output_modalities":["text"],"vendor_name":"Google","vendor_icon":"Gemini.Color","billing_mode":"tiered_expr","billing_expr":"p \u003c= 200000 ? tier(\"standard\", p * 2 + c * 12 + cr * 0.2 + ai * 2) : tier(\"long_context\", p * 4 + c * 18 + cr * 0.4 + ai * 4)","model_group_ratio":{"Economy Mode":0.15,"Tier 1":0.9,"Tier 2":0.85,"Tier 3":0.8,"Tier 4":0.7,"TokenPlan-Official":0.8,"auto":1,"default":1,"group-free":0,"svip":1,"vip":1}}],"group_model_ratio":{"default":{"claude-fable-5":1,"deepseek-v4.1-flash":0.5,"gemini-3.7-flash":0.5,"gemini-3.8-flash":0.5,"glm-5.2":1,"gpt-5.6-sol":1,"gpt-image-2.5-flare":0.9,"gpt-image-2.5-sunburst":0.9,"grok-4.3":1,"jev-1.13.0":0.5,"ling-3.0-flash-fin":0.1,"ling-3.0-flash-sante":0.1,"ling-3.0-flash-vl":0.1}},"group_model_ratio_expiry":{"default":{"deepseek-v4.1-flash":1790179200,"gemini-3.7-flash":1798732740,"gemini-3.8-flash":1798732740,"gpt-image-2.5-flare":1791561600,"gpt-image-2.5-sunburst":1791561600,"jev-1.13.0":1790265600,"ling-3.0-flash-fin":1790092800,"ling-3.0-flash-sante":1790092800,"ling-3.0-flash-vl":1790092800}},"group_ratio":{"Economy Mode":0.15,"auto":1,"default":1,"group-free":0},"pricing_version":"group-model-ratio-v1","success":true,"supported_endpoint":{"anthropic":{"path":"/v1/messages","method":"POST"},"audio":{"path":"/v1/audio/transcriptions","method":"POST"},"embeddings":{"path":"/v1/embeddings","method":"POST"},"gemini":{"path":"/v1beta/models/{model}:generateContent","method":"POST"},"image-generation":{"path":"/v1/images/generations","method":"POST"},"jina-rerank":{"path":"/v1/rerank","method":"POST"},"openai":{"path":"/v1/chat/completions","method":"POST"},"openai-tts":{"path":"/v1/audio/speech","method":"POST"},"openai-video":{"path":"/v1/video/generations","method":"POST"},"suno":{"path":"/suno/submit/lyrics","method":"POST"},"systemone":{"path":"/v1/systemone","method":"POST"}},"usable_group":{"Economy Mode":"Relying on the advantages of channel scale and multi-source computing power integration, the price is better.","auto":"Auto group","default":"Priority guarantee of official resources, SLA：≥ 99.9%， Recommended for use in production environments.","group-free":"Free group cannot guarantee stability"},"vendors":[{"id":2,"name":"Tencent","icon":"Hunyuan.Color"},{"id":23,"name":"Omni","icon":"https://cos.frostai.cn/omnirouters/logo/logo.png"},{"id":40,"name":"上海创智学院","icon":"https://sf-maas.s3.us-east-1.amazonaws.com/Model_LOGO/NEX.svg"},{"id":46,"name":"TypeSafe","icon":"https://cdn.marmot-cloud.com/storage/zenmux/2026/09/20/BFx2T9Y/Property-1Typesafe.svg"},{"id":47,"name":"Convai Innovations","icon":"https://avatars.githubusercontent.com/u/91363996?s=60\u0026v=4"},{"id":13,"name":"Anthropic","icon":"Claude.Color"},{"id":32,"name":"Youdao","icon":"https://sf-maas-uat-prod.oss-cn-shanghai.aliyuncs.com/Model_LOGO/netease-youdao.svg"},{"id":35,"name":"BAAI","icon":"https://sf-maas-uat-prod.oss-cn-shanghai.aliyuncs.com/Model_LOGO/BAAI.svg"},{"id":36,"name":"TeleAI","icon":"\thttps://sf-maas-uat-prod.oss-cn-shanghai.aliyuncs.com/Model_LOGO/TeleAI.svg"},{"id":42,"name":"Vega","icon":"Vega"},{"id":45,"name":"Meta","icon":"Meta"},{"id":1,"name":"Vidu","icon":"Vidu"},{"id":28,"name":"Qwen","icon":"Qwen"},{"id":30,"name":"PixVerse","icon":"PixVerse"},{"id":18,"name":"Z.ai","icon":"https://cdnv2-cache.udelivrs.com/2026/05/acc3a4676ad771bd616079c832b3d7a5_1779342707107.svg"},{"id":20,"name":"Suno","icon":"Suno"},{"id":3,"name":"Dreamina","icon":"Jimeng.Color"},{"id":4,"name":"Kwai","icon":"Kling.Color"},{"id":15,"name":"Midjourney","icon":"Midjourney"},{"id":25,"name":"Moonshot","icon":"Moonshot"},{"id":34,"name":"阿里巴巴","icon":"Qwen.Color"},{"id":44,"name":"讯飞","icon":"Spark.Color"},{"id":16,"name":"DeepSeek","icon":"DeepSeek.Color"},{"id":27,"name":"Kimi","icon":"Kimi"},{"id":33,"name":"智谱","icon":"Zhipu.Color"},{"id":37,"name":"腾讯","icon":"Hunyuan.Color"},{"id":38,"name":"快手","icon":"Kling.Color"},{"id":39,"name":"字节跳动","icon":"Doubao.Color"},{"id":14,"name":"xAI","icon":"https://t0.gstatic.com/faviconV2?client=SOCIAL\u0026type=FAVICON\u0026fallback_opts=TYPE,SIZE,URL\u0026url=https://x.ai/\u0026size=256"},{"id":17,"name":"ByteDance","icon":"https://www-s.ucloud.cn/2025/09/445d859ee13b85ee8715e585991497ed_1758886668504.svg"},{"id":10,"name":"MiniMax","icon":"Minimax"},{"id":43,"name":"InclusionAI","icon":"https://maas-public.oss-cn-shanghai.aliyuncs.com/maas-admin/image/2026-06-18/image-5b9c856a-3b04-40.png"},{"id":5,"name":"OpenAI","icon":"OpenAI"},{"id":6,"name":"Google","icon":"Gemini.Color"},{"id":31,"name":"NVIDIA","icon":"Nvidia"},{"id":41,"name":"Xiaomi","icon":"https://cdnv2-cache.udelivrs.com/2026/05/80caddd717cfafd45b0671c5b8969296_1779343536965.svg"}]}