[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"model-image-service-v2-gpt-5.6-terra":3,"text-service-v2-gpt-5.6-terra":205},[4],{"id":5,"aliases":6,"types":7,"pricing_group":10,"pricing":11,"display_name":5,"model_capabilities":28,"model_capabilities_zh":101,"model_data":150,"introduction":193,"introduction_zh":194,"description":195,"description_zh":196,"title_focus":197,"title_focus_zh":198,"keyword_extras":199,"keyword_extras_zh":202},"gpt-5.6-terra",[5],[8,9],"openai","openai\u002Fresponses\u002Fv1","openai\u002Fgpt",{"input_tokens_tiers":12},{"tiers":13},[14,21],{"max_tokens":15,"pricing":16},272000,{"input_tokens":17,"output_tokens":18,"cache_read_tokens":19,"cache_write_tokens":20},"2","12","0.2","2.5",{"max_tokens":22,"pricing":23},0,{"input_tokens":24,"output_tokens":25,"cache_read_tokens":26,"cache_write_tokens":27},"4","18","0.4","5",{"groups":29,"highlights":90,"overall":94,"scenario":96,"summary":99,"weakness":100},[30,67,80],{"items":31,"name":66},[32,38,43,47,50,54,57,61],{"detail":33,"method":34,"name":35,"samples":36,"score":37},"","Mixed","Agent Collaboration",18,72,{"detail":33,"method":39,"name":40,"samples":41,"score":42},"Objective","Code Generation",12,77,{"detail":33,"method":34,"name":44,"samples":45,"score":46},"Code Comprehension & Debugging",6,85,{"detail":33,"method":34,"name":48,"samples":49,"score":42},"Creative Writing",22,{"detail":33,"method":34,"name":51,"samples":52,"score":53},"Information Retrieval \u002F RAG",9,87,{"detail":33,"method":34,"name":55,"samples":56,"score":46},"Long-Context Handling",16,{"detail":33,"method":34,"name":58,"samples":59,"score":60},"Math & Science Reasoning",14,96,{"detail":33,"method":62,"name":63,"samples":64,"score":65},"Subjective","Multilingual Capability",11,92,"Advanced Capabilities",{"items":68,"name":79},[69,73,77],{"detail":33,"method":34,"name":70,"samples":71,"score":72},"General Knowledge",10,97,{"detail":33,"method":62,"name":74,"samples":75,"score":76},"Text Comprehension",23,80,{"detail":33,"method":34,"name":78,"samples":56,"score":37},"Logical Reasoning","General Foundations",{"items":81,"name":89},[82,86],{"detail":33,"method":34,"name":83,"samples":84,"score":85},"Hallucination Control",15,93,{"detail":33,"method":62,"name":87,"samples":36,"score":88},"Instruction Following",84,"Safety & Compliance",[91,92,93],{"name":70,"score":72},{"name":58,"score":60},{"name":83,"score":85},{"rating":95,"score":88},"良好",{"fit":97,"name":98},86,"Translation & Localization \u002F Customer Support \u002F Enterprise Knowledge Q&A","GPT-5.6 Terra scored 84 overall, performing well — a knowledge and reasoning powerhouse. It stands out in general knowledge (97%), math and science reasoning (96%) and hallucination control (93%), combining encyclopedic coverage with disciplined, low-hallucination output, which makes it dependable for localization, customer support and knowledge-base Q&A. Comparatively, agent collaboration (72%) is moderate; multi-tool workflows need tighter design. Knowledge-dense, accuracy-first tasks are its home turf.",{"name":35,"note":33,"score":37},{"groups":102,"highlights":141,"overall":145,"scenario":146,"summary":148,"weakness":149},[103,125,134],{"items":104,"name":124},[105,108,111,113,115,117,119,121],{"detail":33,"method":106,"name":107,"samples":36,"score":37},"主\u002F客观混合","智能体协同",{"detail":33,"method":109,"name":110,"samples":41,"score":42},"客观","代码生成",{"detail":33,"method":106,"name":112,"samples":45,"score":46},"代码理解与调试",{"detail":33,"method":106,"name":114,"samples":49,"score":42},"创意写作",{"detail":33,"method":106,"name":116,"samples":52,"score":53},"信息检索 \u002F RAG",{"detail":33,"method":106,"name":118,"samples":56,"score":46},"长文本处理",{"detail":33,"method":106,"name":120,"samples":59,"score":60},"数学与科学推理",{"detail":33,"method":122,"name":123,"samples":64,"score":65},"主观","多语言能力","专项进阶能力",{"items":126,"name":133},[127,129,131],{"detail":33,"method":106,"name":128,"samples":71,"score":72},"通识知识",{"detail":33,"method":122,"name":130,"samples":75,"score":76},"文本理解",{"detail":33,"method":106,"name":132,"samples":56,"score":37},"逻辑推理","通用基础能力",{"items":135,"name":140},[136,138],{"detail":33,"method":106,"name":137,"samples":84,"score":85},"幻觉控制",{"detail":33,"method":122,"name":139,"samples":36,"score":88},"指令遵循","安全与合规能力",[142,143,144],{"name":128,"score":72},{"name":120,"score":60},{"name":137,"score":85},{"rating":95,"score":88},{"fit":97,"name":147},"翻译与本地化 \u002F 智能客服 \u002F 企业知识库问答","GPT-5.6 Terra 综合得分 84，整体表现良好，是一款知识与推理型主力模型。通识知识（97%）、数学与科学推理（96%）与幻觉控制（93%）表现突出，百科级知识覆盖叠加低幻觉输出，在翻译本地化与企业知识库问答场景中优势明显。相对而言，智能体协同（72%）一般，多工具工作流需更精心的设计。知识密集、准确性优先的任务是它的主场。",{"name":107,"note":33,"score":37},{"bars":151,"metrics":154,"recentAvailability":181,"summary":182},{"degraded":152,"down":153},[],[],[155,158,161,165,169,173,177],{"label":156,"value":157},"Completion Rate","100%",{"label":159,"value":160},"Cache Hit Rate","0.0%",{"label":162,"unit":163,"value":164},"Total Eval Time","s","2125.2",{"label":166,"unit":167,"value":168},"Token Usage","tokens","159,122",{"label":170,"unit":171,"value":172},"Avg. Requests","次","1.0",{"label":174,"unit":175,"value":176},"TPM","tokens\u002Fmin","4,492",{"label":178,"unit":179,"value":180},"QPM","req\u002Fmin","4.2","100.0%",[183,186,189],{"label":184,"unit":163,"value":185},"Latency","7.8",{"label":187,"unit":175,"value":188},"Throughput","4.49K",{"label":190,"unit":191,"value":192},"Availability","%","100.0","GPT-5.6 Terra is the knowledge-and-reasoning anchor of the GPT-5.6 family, pairing broad accurate knowledge with careful analysis — suited to translation, customer support and enterprise knowledge bases.","GPT-5.6 Terra是GPT-5.6系列的知识与推理主力，兼具广博准确的知识与谨慎的分析能力，适合翻译、智能客服和企业知识库场景。","GPT-5.6 Terra strengthens knowledge coverage, math and science reasoning and factual reliability for knowledge-intensive applications. Integrate in one click via the SilvaMux unified API.","GPT-5.6 Terra强化知识覆盖、数学与科学推理和事实可靠性，适合知识密集型应用。可通过 SilvaMux 统一 API 一键快速接入。","Knowledge & Reasoning Model","知识推理模型",[200,201],"knowledge Q&A","factual reliability",[203,204],"知识问答","事实可靠",{"ttft":206,"tps":209,"e2e":212,"requests":215,"successfulRequests":216,"successRate":217,"availability":218},{"p50":207,"p95":208},1465,2391,{"p50":210,"p95":211},35,66,{"p50":213,"p95":214},4047,7581,104736,102677,"98.03",[219,220,219,219,219,219,219,220,219,219,219,219,219,219,219,219,219,219,219,219,219,219,219,219],"up","degraded"]