{"auto_groups":["claude","gemini","openai","seedance2.0","default"],"data":[{"model_name":"gemini-3.1-flash-lite","description":"Gemini 3.1 Flash Lite is a lightweight tier prioritizing low latency and cost-efficiency, suited to high-concurrency, latency-sensitive workloads.","icon":"Google","vendor_id":2,"quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"","completion_ratio":6,"enable_groups":["gemini"],"supported_endpoint_types":["gemini","openai"],"pricing_version":"5a90f2b86c08bd983a9a2e6d66c255f4eaef9c4bc934386d2b6ae84ef0ff1f1f"},{"model_name":"gemini-3-flash","description":"Gemini 3 Flash is a Google model that pairs speed with strong intelligence, plus capable search and grounding.","icon":"Google","vendor_id":2,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["gemini"],"supported_endpoint_types":["gemini","openai"],"billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 0.5 + ai * 1 + c * 3 + cr * 0.05)"},{"model_name":"gemini-3.1-flash-lite-image","vendor_id":2,"quota_type":0,"model_ratio":0.125,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["gemini","Rezona"],"supported_endpoint_types":["gemini","openai"],"billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 0.25 + c * 1.5 + img_o * 30)"},{"model_name":"gemini-3-pro-preview","vendor_id":2,"quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"","completion_ratio":6,"cache_ratio":0.1,"enable_groups":["gemini"],"supported_endpoint_types":["gemini","openai"]},{"model_name":"gpt-5.6-terra","vendor_id":11,"quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["openai"],"supported_endpoint_types":["openai"]},{"model_name":"gemini-3.5-flash","description":"Gemini 3.5 Flash is Google's most advanced Flash model, delivering consistent, production-ready performance at scale for agentic execution, coding, and long-horizon tasks.","icon":"Google","vendor_id":2,"quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["gemini","Rezona"],"supported_endpoint_types":["gemini","openai"],"billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 1.5 + c * 9 + cr * 0.15)"},{"model_name":"kimi-k3","vendor_id":15,"quota_type":0,"model_ratio":1.477,"model_price":0,"owner_by":"","completion_ratio":12.496615905245347,"cache_ratio":0.1,"enable_groups":["openai"],"supported_endpoint_types":["anthropic","openai"]},{"model_name":"dreamina-seedance-2-0-260128","description":"Seedance 2.0 (Dreamina) is ByteDance's multimodal video generation model. It accepts text, image, audio, and video references in a single pass and produces cinematic, multi-shot video with synchronized native audio, consistent characters, and realistic motion.","vendor_id":12,"quota_type":0,"model_ratio":3.85,"model_price":0,"owner_by":"","completion_ratio":0,"enable_groups":["seedance2.0"],"supported_endpoint_types":["openai-video"]},{"model_name":"dreamina-seedance-2-0-fast-260128","description":"Seedance 2.0 Fast (Dreamina) is the low-latency variant of ByteDance's Seedance 2.0 video model. It keeps the same multimodal text, image, audio, and video inputs while prioritizing faster, lower-cost generation for high-volume and iterative work.","vendor_id":12,"quota_type":0,"model_ratio":2.8,"model_price":0,"owner_by":"","completion_ratio":0,"enable_groups":["seedance2.0"],"supported_endpoint_types":["openai-video"]},{"model_name":"gemini-2.5-flash-thinking","vendor_id":2,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["gemini"],"supported_endpoint_types":["gemini","openai"]},{"model_name":"claude-opus-4-7","description":"Claude Opus 4.7 supports a 1M-token context window, up to 128K output tokens, and adaptive thinking, with the same tool and platform capabilities as Opus 4.6.","icon":"Anthropic","vendor_id":13,"quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["claude"],"supported_endpoint_types":["anthropic","openai"]},{"model_name":"claude-sonnet-5","description":"Claude Sonnet 5 is the most agentic model in the Sonnet line — able to plan, use tools such as browsers and terminals, and operate autonomously at a level once reserved for larger, more expensive models.","icon":"Anthropic","vendor_id":13,"quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["claude"],"supported_endpoint_types":["anthropic","openai"]},{"model_name":"gemini-3.1-flash-image","description":"Gemini 3.1 Flash Image delivers high-quality image generation and conversational editing at mainstream pricing with low latency.","icon":"Google","vendor_id":2,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["gemini","Rezona"],"supported_endpoint_types":["gemini","openai"],"billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 0.5 + c * 3 + img_o * 60)"},{"model_name":"gemini-2.5-flash","description":"Gemini 2.5 Flash is Google's fast, cost-efficient model for high-throughput, latency-sensitive tasks, balancing speed with solid quality.","icon":"Google","vendor_id":2,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["Rezona","gemini"],"supported_endpoint_types":["gemini","openai"],"billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 0.3 + c * 2.5 + cr * 0.03)"},{"model_name":"claude-opus-5","vendor_id":13,"quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["claude"],"supported_endpoint_types":["anthropic","openai"]},{"model_name":"seedance-2-0","description":"Seedance 2.0 is ByteDance's multimodal video generation model. It accepts text, image, audio, and video references in a single pass and produces cinematic, multi-shot video with synchronized native audio, consistent characters, and realistic motion.","vendor_id":12,"quota_type":0,"model_ratio":3.85,"model_price":0,"owner_by":"","completion_ratio":0,"enable_groups":["seedance2.0"],"supported_endpoint_types":["openai-video"]},{"model_name":"seedance-2-0-mini","quota_type":0,"model_ratio":3.85,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["seedance2.0"],"supported_endpoint_types":["openai"]},{"model_name":"grok-4.5","vendor_id":14,"quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"","completion_ratio":3,"cache_ratio":0.25,"enable_groups":["Rezona","openai"],"supported_endpoint_types":["openai-response"]},{"model_name":"claude-fable-5","vendor_id":13,"quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["claude"],"supported_endpoint_types":["anthropic","openai"]},{"model_name":"gemini-2.5-flash-lite","description":"Gemini 2.5 Flash Lite is Google's lightweight, low-cost model with an optional thinking mode, suited to high-volume, latency-sensitive workloads.","icon":"Google","vendor_id":2,"quota_type":0,"model_ratio":0.05,"model_price":0,"owner_by":"","completion_ratio":4,"enable_groups":["gemini"],"supported_endpoint_types":["gemini","openai"]},{"model_name":"gemini-2.5-pro","description":"Gemini 2.5 Pro is a Google reasoning model that excels at coding and complex prompts. Its thinking mode lets it reason before responding for stronger accuracy, with multimodal input across text, images, audio, video, and code.","icon":"Google","vendor_id":2,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["gemini"],"supported_endpoint_types":["gemini","openai"],"billing_mode":"tiered_expr","billing_expr":"len \u003c= 200000                                                                                                                                                                                                                        ? tier(\"standard\", p * 1.25 + c * 10)                                                                                                                                                                                               :tier(\"long_context\", p * 2.5 + c * 15)"},{"model_name":"claude-opus-5-5","vendor_id":13,"quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["claude"],"supported_endpoint_types":["anthropic","openai"]},{"model_name":"gemini-3.1-pro","description":"Gemini 3.1 Pro is a Google model built on advanced reasoning, designed for agentic workflows, autonomous coding, and complex multimodal tasks. It suits complex work that calls for broad world knowledge and advanced cross-modal reasoning.","icon":"Google","vendor_id":2,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["Rezona","gemini"],"supported_endpoint_types":["gemini","openai"],"billing_mode":"tiered_expr","billing_expr":"len \u003c= 200000 ? tier(\"standard\", p * 2 + c * 12 + cr * 0.2) : tier(\"long_context\", p * 4 + c * 18 + cr * 0.4)"},{"model_name":"gemini-2.5-flash-image","description":"Gemini 2.5 Flash Image offers image generation and conversational multi-turn editing, optimized for fast creative workflows.","icon":"Google","vendor_id":2,"quota_type":0,"model_ratio":1.5,"model_price":0,"owner_by":"","completion_ratio":10,"enable_groups":["gemini"],"supported_endpoint_types":["gemini","openai"]},{"model_name":"deepseek-v4-1-flash-260910","vendor_id":17,"quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"","completion_ratio":4,"cache_ratio":0.02,"enable_groups":["openai"],"supported_endpoint_types":["openai"]},{"model_name":"gpt-6-astra","vendor_id":11,"quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"","completion_ratio":5,"enable_groups":["openai"],"supported_endpoint_types":["openai"]},{"model_name":"glm-5.2","quota_type":0,"model_ratio":4,"model_price":0,"owner_by":"","completion_ratio":3.5,"cache_ratio":0.25,"enable_groups":["claude"],"supported_endpoint_types":["openai"]},{"model_name":"seedance-2-0-fast","description":"Seedance 2.0 Fast is the low-latency variant of ByteDance's Seedance 2.0 video model. It keeps the same multimodal text, image, audio, and video inputs while prioritizing faster, lower-cost generation for high-volume and iterative work.","vendor_id":12,"quota_type":0,"model_ratio":2.8,"model_price":0,"owner_by":"","completion_ratio":0,"enable_groups":["seedance2.0"],"supported_endpoint_types":["openai-video"]},{"model_name":"seedance-2-5","description":"Seedance 2.5 is ByteDance's multimodal video generation model. It accepts text, image, audio, and video references in a single pass and produces cinematic, multi-shot video with synchronized native audio, consistent characters, and realistic motion.","vendor_id":12,"quota_type":0,"model_ratio":6.2,"model_price":0,"owner_by":"","completion_ratio":0,"enable_groups":["seedance2.0"],"supported_endpoint_types":["openai-video"]},{"model_name":"deepseek-v4-1-flash","vendor_id":17,"quota_type":0,"model_ratio":0.15,"model_price":0,"owner_by":"","completion_ratio":4,"cache_ratio":0.02,"enable_groups":["openai"],"supported_endpoint_types":["openai"]},{"model_name":"gemini-3-pro-image","description":"Gemini 3 Pro Image (marketed as Nano Banana Pro) is Google's next-generation image generation and editing model, an upgrade to Gemini 2.5 Flash Image (Nano Banana). It combines multimodal Transformer and diffusion approaches, natively supports 2K (2048×2048) and 4K output, and improves image quality, text rendering, and physical reasoning.","icon":"Google","vendor_id":2,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["Rezona","gemini"],"supported_endpoint_types":["gemini","openai"],"billing_mode":"tiered_expr","billing_expr":"tier(\"base\", p * 2 + c * 12 + img_o * 120)"},{"model_name":"gpt-5.4","description":"GPT-5.4 is OpenAI's frontier model for complex professional work.","icon":"OpenAI","vendor_id":11,"quota_type":0,"model_ratio":1.25,"model_price":0,"owner_by":"","completion_ratio":9,"cache_ratio":0.1,"enable_groups":["openai"],"supported_endpoint_types":["openai"]},{"model_name":"gpt-5.5","description":"GPT-5.5 is OpenAI's flagship large language model, released on April 24, 2026. It is designed as a new generation of intelligence for real-world work and AI agents. Its core advancement lies in its ability to autonomously plan and execute complex, multi-step tasks. GPT-5.5 excels in areas such as software engineering, computer operation, and scientific research and analysis, while achieving higher efficiency with lower token consumption.","icon":"OpenAI","quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"","completion_ratio":9,"cache_ratio":0.1,"enable_groups":["openai"],"supported_endpoint_types":["openai"]},{"model_name":"gpt-image-2","description":"GPT Image 2 is OpenAI's most advanced image generation model, supporting fast, high-quality image generation and editing, with flexible image sizes and high-fidelity image input.","icon":"OpenAI","tags":"OpenAI","vendor_id":11,"quota_type":0,"model_ratio":4,"model_price":0,"owner_by":"","completion_ratio":3.75,"cache_ratio":0.25,"image_ratio":1,"enable_groups":["openai"],"supported_endpoint_types":["openai"]},{"model_name":"grok-4.6","vendor_id":14,"quota_type":0,"model_ratio":1,"model_price":0,"owner_by":"","completion_ratio":3,"enable_groups":["Rezona","openai"],"supported_endpoint_types":["openai"]},{"model_name":"gemini-2.5-flash-nothinking","vendor_id":2,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["gemini"],"supported_endpoint_types":["gemini","openai"]},{"model_name":"cluade-sonnet-5","quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["claude"],"supported_endpoint_types":["anthropic","openai"]},{"model_name":"claude-fable-5-1","vendor_id":13,"quota_type":0,"model_ratio":5,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"enable_groups":["claude"],"supported_endpoint_types":["anthropic","openai"]},{"model_name":"gpt-5.6-sol","vendor_id":11,"quota_type":0,"model_ratio":2,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["openai"],"supported_endpoint_types":["openai"]},{"model_name":"gpt-5.6-luna","vendor_id":11,"quota_type":0,"model_ratio":0.1,"model_price":0,"owner_by":"","completion_ratio":6,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["openai"],"supported_endpoint_types":["openai"]},{"model_name":"claude-opus-4-8","description":"Claude Opus 4.8 is Anthropic's most capable Opus model for long-running, complex work — purpose-built for large-scale development, agent building, and scenarios that demand the highest quality and autonomy.","icon":"Anthropic","vendor_id":13,"quota_type":0,"model_ratio":2.5,"model_price":0,"owner_by":"","completion_ratio":5,"cache_ratio":0.1,"create_cache_ratio":1.25,"enable_groups":["claude"],"supported_endpoint_types":["anthropic","openai"]},{"model_name":"gemini-3.7-flash","vendor_id":2,"quota_type":0,"model_ratio":37.5,"model_price":0,"owner_by":"","completion_ratio":1,"enable_groups":["gemini","Rezona"],"supported_endpoint_types":["gemini","openai"],"billing_mode":"tiered_expr","billing_expr":"tier(\"standard\", p * 0.75 + cr * 0.075 + ai * 0.75 + c * 3.75)"}],"group_ratio":{"Rezona":1,"auto":1,"claude":1,"gemini":1,"openai":1,"seedance2.0":1},"pricing_version":"a42d372ccf0b5dd13ecf71203521f9d2","success":true,"supported_endpoint":{"anthropic":{"path":"/v1/messages","method":"POST"},"gemini":{"path":"/v1beta/models/{model}:generateContent","method":"POST"},"openai":{"path":"/v1/chat/completions","method":"POST"},"openai-response":{"path":"/v1/responses","method":"POST"},"openai-video":{"path":"/v1/video/generations","method":"POST"}},"usable_group":{"Rezona":"","auto":"","claude":"","gemini":"","openai":"","seedance2.0":""},"vendors":[{"id":4,"name":"QiHang"},{"id":6,"name":"NanoGPT"},{"id":16,"name":"讯飞","icon":"Spark.Color"},{"id":7,"name":"Vertex","icon":"VertexAI.Color"},{"id":8,"name":"Qiniu"},{"id":13,"name":"Anthropic","icon":"Claude.Color"},{"id":17,"name":"DeepSeek","icon":"DeepSeek.Color"},{"id":14,"name":"xAI","icon":"XAI"},{"id":1,"name":"SAP AI Core"},{"id":5,"name":"OpenCode Zen","icon":"OpenCode"},{"id":9,"name":"Venice AI","icon":"VeniceAI"},{"id":10,"name":"LLM Gateway"},{"id":15,"name":"Moonshot","icon":"Moonshot"},{"id":11,"name":"OpenAI","icon":"OpenAI"},{"id":12,"name":"字节跳动","icon":"Doubao.Color"},{"id":2,"name":"Google","icon":"Gemini.Color"},{"id":3,"name":"Vivgrid"}]}