[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"model-image-service-v2-gemini-3.1-pro-preview":3,"text-service-v2-gemini-3.1-pro-preview":205},[4],{"id":5,"aliases":6,"types":7,"pricing_group":9,"pricing":10,"display_name":5,"capabilities":25,"model_capabilities":30,"model_capabilities_zh":103,"introduction":152,"introduction_zh":153,"description":154,"description_zh":155,"title_focus":156,"title_focus_zh":157,"keyword_extras":158,"keyword_extras_zh":161,"model_data":164},"gemini-3.1-pro-preview",[5],[8],"openai","google\u002Fgemini",{"input_tokens_tiers":11},{"tiers":12},[13,19],{"max_tokens":14,"pricing":15},200000,{"input_tokens":16,"output_tokens":17,"cache_read_tokens":18,"cache_write_tokens":16},"2","12","0.2",{"max_tokens":20,"pricing":21},0,{"input_tokens":22,"output_tokens":23,"cache_read_tokens":24,"cache_write_tokens":22},"4","18","0.4",{"context_window":26,"max_output":27,"modalities":28},1048576,65536,[29],"text",{"groups":31,"highlights":92,"overall":96,"scenario":98,"summary":101,"weakness":102},[32,70,83],{"items":33,"name":69},[34,40,45,49,54,58,62,65],{"detail":35,"method":36,"name":37,"samples":38,"score":39},"","Mixed","Agent Collaboration",18,59,{"detail":35,"method":41,"name":42,"samples":43,"score":44},"Objective","Code Generation",12,42,{"detail":35,"method":36,"name":46,"samples":47,"score":48},"Code Comprehension & Debugging",6,40,{"detail":35,"method":50,"name":51,"samples":52,"score":53},"Subjective","Creative Writing",22,77,{"detail":35,"method":36,"name":55,"samples":56,"score":57},"Information Retrieval \u002F RAG",9,74,{"detail":35,"method":50,"name":59,"samples":60,"score":61},"Long-Context Handling",16,83,{"detail":35,"method":41,"name":63,"samples":64,"score":53},"Math & Science Reasoning",14,{"detail":35,"method":50,"name":66,"samples":67,"score":68},"Multilingual Capability",11,89,"Advanced Capabilities",{"items":71,"name":82},[72,76,80],{"detail":35,"method":36,"name":73,"samples":74,"score":75},"General Knowledge",10,92,{"detail":35,"method":50,"name":77,"samples":78,"score":79},"Text Comprehension",23,80,{"detail":35,"method":36,"name":81,"samples":60,"score":57},"Logical Reasoning","General Foundations",{"items":84,"name":91},[85,88],{"detail":35,"method":36,"name":86,"samples":87,"score":61},"Hallucination Control",15,{"detail":35,"method":50,"name":89,"samples":38,"score":90},"Instruction Following",87,"Safety & Compliance",[93,94,95],{"name":73,"score":75},{"name":66,"score":68},{"name":89,"score":90},{"rating":97,"score":57},"良好",{"fit":99,"name":100},84,"Translation & Localization \u002F Scriptwriting \u002F E-commerce Copywriting","Gemini 3.1 Pro scored 74 overall, performing well — a content-oriented professional model. It stands out in general knowledge (92%), multilingual capability (89%) and instruction following (87%), producing well-structured multilingual content to spec, which suits translation, scriptwriting and e-commerce copy. Comparatively, code comprehension and debugging (40%) is markedly weak — coding tasks are effectively a blind spot. Structured, multilingual content work is where it performs.",{"name":46,"note":35,"score":48},{"groups":104,"highlights":143,"overall":147,"scenario":148,"summary":150,"weakness":151},[105,127,136],{"items":106,"name":126},[107,110,113,115,118,120,122,124],{"detail":35,"method":108,"name":109,"samples":38,"score":39},"主\u002F客观混合","智能体协同",{"detail":35,"method":111,"name":112,"samples":43,"score":44},"客观","代码生成",{"detail":35,"method":108,"name":114,"samples":47,"score":48},"代码理解与调试",{"detail":35,"method":116,"name":117,"samples":52,"score":53},"主观","创意写作",{"detail":35,"method":108,"name":119,"samples":56,"score":57},"信息检索 \u002F RAG",{"detail":35,"method":116,"name":121,"samples":60,"score":61},"长文本处理",{"detail":35,"method":111,"name":123,"samples":64,"score":53},"数学与科学推理",{"detail":35,"method":116,"name":125,"samples":67,"score":68},"多语言能力","专项进阶能力",{"items":128,"name":135},[129,131,133],{"detail":35,"method":108,"name":130,"samples":74,"score":75},"通识知识",{"detail":35,"method":116,"name":132,"samples":78,"score":79},"文本理解",{"detail":35,"method":108,"name":134,"samples":60,"score":57},"逻辑推理","通用基础能力",{"items":137,"name":142},[138,140],{"detail":35,"method":108,"name":139,"samples":87,"score":61},"幻觉控制",{"detail":35,"method":116,"name":141,"samples":38,"score":90},"指令遵循","安全与合规能力",[144,145,146],{"name":130,"score":75},{"name":125,"score":68},{"name":141,"score":90},{"rating":97,"score":57},{"fit":99,"name":149},"翻译与本地化 \u002F 短剧剧本生成 \u002F 电商营销文案","Gemini 3.1 Pro 综合得分 74，整体表现良好，是一款内容导向的专业模型。通识知识（92%）、多语言能力（89%）与指令遵循（87%）表现突出，能按规格产出结构化多语言内容，在翻译本地化与电商营销文案场景中优势明显。相对而言，代码理解与调试（40%）明显不足，代码类任务基本是盲区。结构化多语言内容工作是它的舞台。",{"name":114,"note":35,"score":48},"Gemini 3.1 Pro is Google's new professional model with broad knowledge and multilingual strength, handling long-form and structured content — suited to localization, scriptwriting and e-commerce copywriting.","Gemini 3.1 Pro是Google的新一代专业模型，通识知识与多语言能力全面，擅长长文本与结构化内容处理，适合翻译本地化、短剧剧本生成和电商营销文案。","Gemini 3.1 Pro strengthens knowledge understanding and multilingual generation for professional content and long-form tasks. Integrate in one click via the SilvaMux unified API.","Gemini 3.1 Pro强化知识理解与多语言生成，适合专业内容创作与长文本任务。可通过 SilvaMux 统一 API 一键快速接入。","Professional Content Model","专业内容模型",[159,160],"multilingual","long-form content",[162,163],"多语言","长文本",{"summary":165,"recentAvailability":178,"observedAt":35,"bars":179,"metrics":183},[166,170,174],{"label":167,"value":168,"unit":169},"modelsDetail.performance.latency","50.59","s",{"label":171,"value":172,"unit":173},"modelsDetail.performance.throughput","98.90K","tokens\u002Fmin",{"label":175,"value":176,"unit":177},"modelsDetail.performance.availability","98.8","%","98.8%",{"degraded":180,"down":181},[],[182,39],56,[184,187,190,194,198,201],{"label":185,"value":186},"modelsDetail.performance.completionRate","98%",{"label":188,"value":189},"modelsDetail.performance.cacheHitRate","82%",{"label":191,"value":192,"unit":193},"modelsDetail.performance.tokenUsage","1,500,517,319","tokens",{"label":195,"value":196,"unit":197},"modelsDetail.performance.avgRequests","48","modelsDetail.performance.requestsUnit",{"label":199,"value":200,"unit":173},"modelsDetail.performance.tpm","98,901",{"label":202,"value":203,"unit":204},"modelsDetail.performance.qpm","566","req\u002Fmin",{"ttft":206,"tps":209,"e2e":212,"requests":215,"successfulRequests":216,"successRate":217,"availability":218},{"p50":207,"p95":208},1992,2049,{"p50":210,"p95":211},36,75,{"p50":213,"p95":214},6004,7474,194578,192706,"99.04",[219,219,219,219,219,219,219,219,219,219,219,219,219,220,219,219,219,219,219,219,219,219,219,219],"up","degraded"]