[{"data":1,"prerenderedAt":813},["ShallowReactive",2],{"navigation":3,"\u002Fmodels\u002Feval-template":312,"\u002Fmodels\u002Feval-template-surround":808},[4,21,54,95,153,169,188,251,280,292],{"title":5,"path":6,"stem":7,"children":8,"icon":20},"快速开始","\u002Fstart","1.start\u002F0.index",[9,11,15],{"title":5,"path":6,"stem":7,"icon":10},"i-lucide-house",{"title":12,"path":13,"stem":14,"icon":10},"使用 TokenResearch Docs","\u002Fstart\u002Felysia-code","1.start\u002F1.elysia-code",{"title":16,"path":17,"stem":18,"icon":19},"安装Node.js 和 Git","\u002Fstart\u002Fnode-git","1.start\u002F2.node-git","i-simple-icons-javascript",false,{"title":22,"path":23,"stem":24,"children":25,"icon":20},"基础教程","\u002Ftutorials","2.tutorials\u002F1.index",[26,29,34,39,44,49],{"title":27,"path":23,"stem":24,"icon":28},"教程路线图","i-lucide-route",{"title":30,"path":31,"stem":32,"icon":33},"环境准备","\u002Ftutorials\u002Fenvironment","2.tutorials\u002F2.environment","i-lucide-settings-2",{"title":35,"path":36,"stem":37,"icon":38},"Codex 首次运行","\u002Ftutorials\u002Fcodex-first-run","2.tutorials\u002F3.codex-first-run","i-lucide-terminal",{"title":40,"path":41,"stem":42,"icon":43},"Agent 规则","\u002Ftutorials\u002Fagent-rules","2.tutorials\u002F4.agent-rules","i-lucide-scroll-text",{"title":45,"path":46,"stem":47,"icon":48},"视觉验收","\u002Ftutorials\u002Fvisual-review","2.tutorials\u002F5.visual-review","i-lucide-monitor-check",{"title":50,"path":51,"stem":52,"icon":53},"质量门禁","\u002Ftutorials\u002Fquality-gates","2.tutorials\u002F6.quality-gates","i-lucide-shield-check",{"title":55,"path":56,"stem":57,"children":58,"icon":20},"Codex 配置文档","\u002Fcodex","3.codex\u002F0.index",[59,61,66,71,76,80,85,90],{"title":55,"path":56,"stem":57,"icon":60},"i-lucide-bot",{"title":62,"path":63,"stem":64,"icon":65},"安装Codex CLI","\u002Fcodex\u002Finstall-cli","3.codex\u002F1.install-cli","i-simple-icons-openai",{"title":67,"path":68,"stem":69,"icon":70},"安装 IDE 插件","\u002Fcodex\u002Finstall-ide","3.codex\u002F2.install-ide","i-lucide-puzzle",{"title":72,"path":73,"stem":74,"icon":75},"Codex App桌面端","\u002Fcodex\u002Fapp-desktop","3.codex\u002F3.app-desktop","i-lucide-flask-conical",{"title":77,"path":78,"stem":79,"icon":60},"Codex配置\u002F使用","\u002Fcodex\u002Fmodel","3.codex\u002F5.model",{"title":81,"path":82,"stem":83,"icon":84},"其他错误","\u002Fcodex\u002Fother-errors","3.codex\u002F6.other-errors","i-lucide-info",{"title":86,"path":87,"stem":88,"icon":89},"网络与代理","\u002Fcodex\u002Fnetwork-proxy","3.codex\u002F7.network-proxy","i-lucide-globe",{"title":91,"path":92,"stem":93,"icon":94},"Codex模型测试","\u002Fcodex\u002Fmodel-test","3.codex\u002F8.model-test","i-lucide-boxes",{"title":96,"path":97,"stem":98,"children":99,"icon":20},"模型中心","\u002Fmodels","4.models\u002F01.index",[100,102,106,111,116,121,125,129,134,139,144,149],{"title":96,"path":97,"stem":98,"icon":101},"i-lucide-store",{"title":103,"path":104,"stem":105,"icon":28},"任务路由建议","\u002Fmodels\u002Ftask-routing","4.models\u002F02.task-routing",{"title":107,"path":108,"stem":109,"icon":110},"推理等级评测","\u002Fmodels\u002Freasoning-levels","4.models\u002F03.reasoning-levels","i-lucide-brain",{"title":112,"path":113,"stem":114,"icon":115},"上下文与压缩","\u002Fmodels\u002Fcontext-compression","4.models\u002F04.context-compression","i-lucide-file-stack",{"title":117,"path":118,"stem":119,"icon":120},"评测记录模板","\u002Fmodels\u002Feval-template","4.models\u002F05.eval-template","i-lucide-clipboard-list",{"title":122,"path":123,"stem":124,"icon":48},"前端视觉评测","\u002Fmodels\u002Ffrontend-visual","4.models\u002F06.frontend-visual",{"title":126,"path":127,"stem":128,"icon":43},"研究写作评测","\u002Fmodels\u002Fresearch-writing","4.models\u002F07.research-writing",{"title":130,"path":131,"stem":132,"icon":133},"运维排查评测","\u002Fmodels\u002Fops-debug","4.models\u002F08.ops-debug","i-lucide-server-cog",{"title":135,"path":136,"stem":137,"icon":138},"成本与速度评测","\u002Fmodels\u002Fcost-speed","4.models\u002F09.cost-speed","i-lucide-timer",{"title":140,"path":141,"stem":142,"icon":143},"工具调用评测","\u002Fmodels\u002Ftool-use","4.models\u002F10.tool-use","i-lucide-terminal-square",{"title":145,"path":146,"stem":147,"icon":148},"样例评测记录","\u002Fmodels\u002Fsample-records","4.models\u002F11.sample-records","i-lucide-clipboard-check",{"title":150,"path":151,"stem":152,"icon":101},"模型广场","\u002Fmodels\u002Fmarket","4.models\u002F12.market",{"title":154,"path":155,"stem":156,"children":157,"icon":20},"OpenCode 配置文档","\u002Fopencode","5.opencode\u002F0.index",[158,160,165],{"title":154,"path":155,"stem":156,"icon":159},"i-lucide-code-xml",{"title":161,"path":162,"stem":163,"icon":164},"安装OpenCode","\u002Fopencode\u002Finstall-opencode","5.opencode\u002F1.install-opencode","i-lucide-code",{"title":166,"path":167,"stem":168,"icon":164},"配置模型","\u002Fopencode\u002Fmodel","5.opencode\u002F2.model",{"title":170,"path":171,"stem":172,"children":173,"icon":20},"Claude Code 配置文档","\u002Fclaude-code","6.claude-code\u002F0.index",[174,176,180,184],{"title":170,"path":171,"stem":172,"icon":175},"i-lucide-sparkles",{"title":177,"path":178,"stem":179,"icon":175},"安装Claude CLI","\u002Fclaude-code\u002Finstall-cli","6.claude-code\u002F1.install-cli",{"title":181,"path":182,"stem":183,"icon":70},"插件配置","\u002Fclaude-code\u002Fplugin","6.claude-code\u002F2.plugin",{"title":185,"path":186,"stem":187,"icon":94},"模型配置","\u002Fclaude-code\u002Fmodel","6.claude-code\u002F3.model",{"title":189,"path":190,"stem":191,"children":192,"icon":20},"Skill Gallery","\u002Fgallery","7.gallery\u002F01.index",[193,195,200,205,210,215,219,224,229,234,238,242,247],{"title":189,"path":190,"stem":191,"icon":194},"i-lucide-gallery-horizontal-end",{"title":196,"path":197,"stem":198,"icon":199},"Image Generation","\u002Fgallery\u002Fimagegen","7.gallery\u002F02.imagegen","i-lucide-image",{"title":201,"path":202,"stem":203,"icon":204},"Paper Analyze","\u002Fgallery\u002Fpaper-analyze","7.gallery\u002F03.paper-analyze","i-lucide-file-search",{"title":206,"path":207,"stem":208,"icon":209},"Presentation Builder","\u002Fgallery\u002Fpresentation","7.gallery\u002F04.presentation","i-lucide-presentation",{"title":211,"path":212,"stem":213,"icon":214},"Spreadsheet Analyst","\u002Fgallery\u002Fspreadsheet","7.gallery\u002F05.spreadsheet","i-lucide-table-properties",{"title":216,"path":217,"stem":218,"icon":48},"Browser Visual QA","\u002Fgallery\u002Fbrowser-visual-qa","7.gallery\u002F06.browser-visual-qa",{"title":220,"path":221,"stem":222,"icon":223},"Zotero Research","\u002Fgallery\u002Fzotero-research","7.gallery\u002F07.zotero-research","i-lucide-library",{"title":225,"path":226,"stem":227,"icon":228},"Research Loop","\u002Fgallery\u002Fresearch-loop","7.gallery\u002F08.research-loop","i-lucide-orbit",{"title":230,"path":231,"stem":232,"icon":233},"PDF and Docs","\u002Fgallery\u002Fpdf-docs","7.gallery\u002F09.pdf-docs","i-lucide-file-type",{"title":235,"path":236,"stem":237,"icon":133},"Ops Debug","\u002Fgallery\u002Fops-debug","7.gallery\u002F10.ops-debug",{"title":239,"path":240,"stem":241,"icon":43},"LaTeX Paper","\u002Fgallery\u002Flatex-paper","7.gallery\u002F11.latex-paper",{"title":243,"path":244,"stem":245,"icon":246},"Diagram Canvas","\u002Fgallery\u002Fdiagram-canvas","7.gallery\u002F12.diagram-canvas","i-lucide-git-compare-arrows",{"title":248,"path":249,"stem":250,"icon":175},"Hatch Pet","\u002Fgallery\u002Fhatch-pet","7.gallery\u002F13.hatch-pet",{"title":252,"path":253,"stem":254,"children":255,"icon":20},"场景路线","\u002Fworkflows","8.workflows\u002F01.index",[256,258,263,268,272,276],{"title":252,"path":253,"stem":254,"icon":257},"i-lucide-map",{"title":259,"path":260,"stem":261,"icon":262},"论文到汇报","\u002Fworkflows\u002Fpaper-to-slides","8.workflows\u002F02.paper-to-slides","i-lucide-file-chart-column",{"title":264,"path":265,"stem":266,"icon":267},"视觉复刻路线","\u002Fworkflows\u002Fvisual-replica","8.workflows\u002F03.visual-replica","i-lucide-monitor-dot",{"title":269,"path":270,"stem":271,"icon":28},"模型路由记录","\u002Fworkflows\u002Fmodel-routing","8.workflows\u002F04.model-routing",{"title":273,"path":274,"stem":275,"icon":194},"Skill 沉淀路线","\u002Fworkflows\u002Fskill-to-gallery","8.workflows\u002F05.skill-to-gallery",{"title":277,"path":278,"stem":279,"icon":133},"运维现场排查","\u002Fworkflows\u002Fops-live-debug","8.workflows\u002F06.ops-live-debug",{"title":281,"path":282,"stem":283,"children":284,"icon":20},"常见问题","\u002Ffaq","91.faq\u002F0.index",[285,288],{"title":286,"path":282,"stem":283,"icon":287},"FAQ","i-lucide-circle-help",{"title":289,"path":290,"stem":291,"icon":84},"通用问题","\u002Ffaq\u002Fcommon","91.faq\u002F1.common",{"title":293,"path":294,"stem":295,"children":296,"icon":20},"Skills 与 MCP","\u002Fskills-mcp","92.skills-mcp\u002F0.index",[297,299,304,308],{"title":293,"path":294,"stem":295,"icon":298},"i-lucide-plug",{"title":300,"path":301,"stem":302,"icon":303},"Skills 推荐","\u002Fskills-mcp\u002Fskills","92.skills-mcp\u002F1.skills","i-lucide-link",{"title":305,"path":306,"stem":307,"icon":298},"MCP 使用边界","\u002Fskills-mcp\u002Fmcp-overview","92.skills-mcp\u002F2.mcp-overview",{"title":309,"path":310,"stem":311,"icon":303},"chrome-devtools-mcp","\u002Fskills-mcp\u002Fchrome-devtools-mcp","92.skills-mcp\u002F3.chrome-devtools-mcp",{"id":313,"title":117,"body":314,"description":802,"extension":326,"links":803,"meta":804,"navigation":805,"path":118,"seo":806,"stem":119,"__hash__":807},"docs\u002F4.models\u002F05.eval-template.md",{"type":315,"value":316,"toc":795},"minimark",[317,321,631,634,638,641,693,696,774,781,784,791],[318,319,320],"h2",{"id":320},"模板",[322,323,328],"pre",{"className":324,"code":325,"language":326,"meta":327,"style":327},"language-md shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","---\ntitle: 评测标题\ntool: Codex Desktop | Codex CLI | OpenCode | Claude Code\nprovider: provider-name\nmodel: 模型名\nreasoning: low | medium | high | xhigh\ndate: YYYY-MM-DD\n---\n\n## 任务\n\n说明输入、目标和约束。\n\n## 过程\n\n- 工具调用：\n- 运行命令：\n- 关键错误：\n- 修复轮数：\n- 开始时间：\n- 结束时间：\n- 总尝试次数：\n- 工作目录：\n- 变更文件：\n\n## 结果\n\n- 是否通过 lint：\n- 是否通过 typecheck：\n- 是否通过 build：\n- 是否通过 route check：\n- 是否通过视觉检查：\n- 关键截图：\n- 关键产物：\n- 命令输出摘要：\n- artifact 路径：\n- screenshot 路径：\n- 未验证风险：\n\n## 结论\n\n这个模型适合什么，不适合什么。\n","md","",[329,330,331,340,346,352,358,364,370,376,382,389,399,404,410,415,423,428,437,445,453,461,469,477,485,493,501,506,514,519,527,535,543,551,559,567,575,583,591,599,607,612,620,625],"code",{"__ignoreMap":327},[332,333,336],"span",{"class":334,"line":335},"line",1,[332,337,339],{"class":338},"sTEyZ","---\n",[332,341,343],{"class":334,"line":342},2,[332,344,345],{"class":338},"title: 评测标题\n",[332,347,349],{"class":334,"line":348},3,[332,350,351],{"class":338},"tool: Codex Desktop | Codex CLI | OpenCode | Claude Code\n",[332,353,355],{"class":334,"line":354},4,[332,356,357],{"class":338},"provider: provider-name\n",[332,359,361],{"class":334,"line":360},5,[332,362,363],{"class":338},"model: 模型名\n",[332,365,367],{"class":334,"line":366},6,[332,368,369],{"class":338},"reasoning: low | medium | high | xhigh\n",[332,371,373],{"class":334,"line":372},7,[332,374,375],{"class":338},"date: YYYY-MM-DD\n",[332,377,379],{"class":334,"line":378},8,[332,380,339],{"class":381},"sMK4o",[332,383,385],{"class":334,"line":384},9,[332,386,388],{"emptyLinePlaceholder":387},true,"\n",[332,390,392,395],{"class":334,"line":391},10,[332,393,394],{"class":381},"## ",[332,396,398],{"class":397},"sBMFI","任务\n",[332,400,402],{"class":334,"line":401},11,[332,403,388],{"emptyLinePlaceholder":387},[332,405,407],{"class":334,"line":406},12,[332,408,409],{"class":338},"说明输入、目标和约束。\n",[332,411,413],{"class":334,"line":412},13,[332,414,388],{"emptyLinePlaceholder":387},[332,416,418,420],{"class":334,"line":417},14,[332,419,394],{"class":381},[332,421,422],{"class":397},"过程\n",[332,424,426],{"class":334,"line":425},15,[332,427,388],{"emptyLinePlaceholder":387},[332,429,431,434],{"class":334,"line":430},16,[332,432,433],{"class":381},"-",[332,435,436],{"class":338}," 工具调用：\n",[332,438,440,442],{"class":334,"line":439},17,[332,441,433],{"class":381},[332,443,444],{"class":338}," 运行命令：\n",[332,446,448,450],{"class":334,"line":447},18,[332,449,433],{"class":381},[332,451,452],{"class":338}," 关键错误：\n",[332,454,456,458],{"class":334,"line":455},19,[332,457,433],{"class":381},[332,459,460],{"class":338}," 修复轮数：\n",[332,462,464,466],{"class":334,"line":463},20,[332,465,433],{"class":381},[332,467,468],{"class":338}," 开始时间：\n",[332,470,472,474],{"class":334,"line":471},21,[332,473,433],{"class":381},[332,475,476],{"class":338}," 结束时间：\n",[332,478,480,482],{"class":334,"line":479},22,[332,481,433],{"class":381},[332,483,484],{"class":338}," 总尝试次数：\n",[332,486,488,490],{"class":334,"line":487},23,[332,489,433],{"class":381},[332,491,492],{"class":338}," 工作目录：\n",[332,494,496,498],{"class":334,"line":495},24,[332,497,433],{"class":381},[332,499,500],{"class":338}," 变更文件：\n",[332,502,504],{"class":334,"line":503},25,[332,505,388],{"emptyLinePlaceholder":387},[332,507,509,511],{"class":334,"line":508},26,[332,510,394],{"class":381},[332,512,513],{"class":397},"结果\n",[332,515,517],{"class":334,"line":516},27,[332,518,388],{"emptyLinePlaceholder":387},[332,520,522,524],{"class":334,"line":521},28,[332,523,433],{"class":381},[332,525,526],{"class":338}," 是否通过 lint：\n",[332,528,530,532],{"class":334,"line":529},29,[332,531,433],{"class":381},[332,533,534],{"class":338}," 是否通过 typecheck：\n",[332,536,538,540],{"class":334,"line":537},30,[332,539,433],{"class":381},[332,541,542],{"class":338}," 是否通过 build：\n",[332,544,546,548],{"class":334,"line":545},31,[332,547,433],{"class":381},[332,549,550],{"class":338}," 是否通过 route check：\n",[332,552,554,556],{"class":334,"line":553},32,[332,555,433],{"class":381},[332,557,558],{"class":338}," 是否通过视觉检查：\n",[332,560,562,564],{"class":334,"line":561},33,[332,563,433],{"class":381},[332,565,566],{"class":338}," 关键截图：\n",[332,568,570,572],{"class":334,"line":569},34,[332,571,433],{"class":381},[332,573,574],{"class":338}," 关键产物：\n",[332,576,578,580],{"class":334,"line":577},35,[332,579,433],{"class":381},[332,581,582],{"class":338}," 命令输出摘要：\n",[332,584,586,588],{"class":334,"line":585},36,[332,587,433],{"class":381},[332,589,590],{"class":338}," artifact 路径：\n",[332,592,594,596],{"class":334,"line":593},37,[332,595,433],{"class":381},[332,597,598],{"class":338}," screenshot 路径：\n",[332,600,602,604],{"class":334,"line":601},38,[332,603,433],{"class":381},[332,605,606],{"class":338}," 未验证风险：\n",[332,608,610],{"class":334,"line":609},39,[332,611,388],{"emptyLinePlaceholder":387},[332,613,615,617],{"class":334,"line":614},40,[332,616,394],{"class":381},[332,618,619],{"class":397},"结论\n",[332,621,623],{"class":334,"line":622},41,[332,624,388],{"emptyLinePlaceholder":387},[332,626,628],{"class":334,"line":627},42,[332,629,630],{"class":338},"这个模型适合什么，不适合什么。\n",[318,632,633],{"id":633},"对比方法",[635,636,637],"p",{},"每次只改变一个变量：模型、推理等级、上下文大小或压缩策略。否则结论很难复用。",[318,639,640],{"id":640},"必须记录的证据",[642,643,644,657],"table",{},[645,646,647],"thead",{},[648,649,650,654],"tr",{},[651,652,653],"th",{},"任务类型",[651,655,656],{},"必须证据",[658,659,660,669,677,685],"tbody",{},[648,661,662,666],{},[663,664,665],"td",{},"代码修改",[663,667,668],{},"lint、typecheck、build 或测试输出",[648,670,671,674],{},[663,672,673],{},"前端视觉",[663,675,676],{},"桌面、移动、暗色截图",[648,678,679,682],{},[663,680,681],{},"文档\u002F论文",[663,683,684],{},"来源、引用、局限性和可复查文件",[648,686,687,690],{},[663,688,689],{},"运维排查",[663,691,692],{},"配置读回、服务状态、日志和客户端探测",[318,694,695],{"id":695},"字段说明",[642,697,698,708],{},[645,699,700],{},[648,701,702,705],{},[651,703,704],{},"字段",[651,706,707],{},"必填原因",[658,709,710,718,726,734,742,750,758,766],{},[648,711,712,715],{},[663,713,714],{},"tool \u002F provider \u002F model",[663,716,717],{},"同一个模型在不同工具表面表现可能不同",[648,719,720,723],{},[663,721,722],{},"reasoning",[663,724,725],{},"用于判断是否该升级或降级推理等级",[648,727,728,731],{},[663,729,730],{},"开始时间 \u002F 结束时间",[663,732,733],{},"记录耗时，避免只看最终质量",[648,735,736,739],{},[663,737,738],{},"总尝试次数",[663,740,741],{},"反映返工成本",[648,743,744,747],{},[663,745,746],{},"运行命令",[663,748,749],{},"让别人能复现验证",[648,751,752,755],{},[663,753,754],{},"artifact 路径",[663,756,757],{},"指向最终文件、日志、PDF、PPTX 或 build 输出",[648,759,760,763],{},[663,761,762],{},"screenshot 路径",[663,764,765],{},"前端和文档渲染任务必须留截图证据",[648,767,768,771],{},[663,769,770],{},"未验证风险",[663,772,773],{},"防止把“本地通过”误写成“生产可用”",[635,775,776,777,780],{},"可以参考 ",[778,779,145],"a",{"href":146},"，不要只写“感觉更好”。",[318,782,783],{"id":783},"命名建议",[635,785,786,787,790],{},"评测记录标题建议包含任务、模型和日期，例如 ",[329,788,789],{},"frontend-visual-deep-high-2026-07-09","。如果同一任务比较多个模型，保持输入、命令和截图视口一致。",[792,793,794],"style",{},"html pre.shiki code .sTEyZ, html code.shiki .sTEyZ{--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8}html pre.shiki code .sMK4o, html code.shiki .sMK4o{--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF}html pre.shiki code .sBMFI, html code.shiki .sBMFI{--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B}html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"title":327,"searchDepth":335,"depth":342,"links":796},[797,798,799,800,801],{"id":320,"depth":342,"text":320},{"id":633,"depth":342,"text":633},{"id":640,"depth":342,"text":640},{"id":695,"depth":342,"text":695},{"id":783,"depth":342,"text":783},"统一记录模型评测结果，方便后续比较。",null,{},{"icon":120},{"title":117,"description":802},"W1uuHnAaJXLwpbCpoXboCGSFqAovgU0YLE09J1dsTBM",[809,811],{"title":112,"path":113,"stem":114,"description":810,"icon":115,"children":-1},"记录长上下文、压缩和 1M 上下文的使用边界。",{"title":122,"path":123,"stem":124,"description":812,"icon":48,"children":-1},"评估模型是否能根据真实浏览器截图修复 UI 问题。",1783590614021]