{ "name": "Model Atlas 模型图谱", "description": "自动维护的 AI 模型资料库和公开可核验真实使用案例库。", "datasetCut": "2026-07-27", "methodology": "Model Atlas 2.0 证据策略", "metrics": { "vendors": 17, "models": 134, "bitableModels": 107, "vendorPageOnlyModels": 12, "publishableModels": 116, "limitedModels": 0, "archiveModels": 2, "holdModels": 16, "verifiedACases": 680, "modelsWithCases": 116, "modelsWithoutCases": 18, "modelsWithSupplementalSources": 119, "modelsAAOnly": 0, "datasetCut": "2026-07-27", "methodology": "Model Atlas 2.0 证据策略", "visualDirection": "Signal Observatory x Living Archive", "sourceNote": "Feishu Bitable sync + official patch preservation + automatic evidence gate; missing fields remain explicit.", "minCasesPerModel": 3, "targetCasesPerModel": 5, "modelsMeetingMinCaseCoverage": 116, "modelsMeetingTargetCaseCoverage": 116, "caseDeficitToMin": 54, "caseDeficitToTarget": 90, "activeModels": 116, "activeModelsWithCases": 116, "activeModelsWithoutCases": 0, "activeModelsMeetingMinCaseCoverage": 116, "activeModelsMeetingTargetCaseCoverage": 116, "activeCaseDeficitToMin": 0, "activeCaseDeficitToTarget": 0, "officialPatchModels": 1 }, "routes": { "home": "https://model-seven-blond.vercel.app/", "models": "https://model-seven-blond.vercel.app/models", "cases": "https://model-seven-blond.vercel.app/cases", "compare": "https://model-seven-blond.vercel.app/compare", "updates": "https://model-seven-blond.vercel.app/updates", "codingAgent": "https://model-seven-blond.vercel.app/topics/coding-agent", "methodology": "https://model-seven-blond.vercel.app/methodology", "commandIndex": "https://model-seven-blond.vercel.app/command-index.json", "evidenceArchive": "https://model-seven-blond.vercel.app/evidence-archive.json", "evidenceArchiveHistory": "https://model-seven-blond.vercel.app/evidence-archive-history.json", "llms": "https://model-seven-blond.vercel.app/llms.txt", "sitemap": "https://model-seven-blond.vercel.app/sitemap.xml" }, "counts": { "vendors": 17, "models": 134, "activeModels": 116, "aCases": 680 }, "evidenceArchive": { "url": "https://model-seven-blond.vercel.app/evidence-archive.json", "historyUrl": "https://model-seven-blond.vercel.app/evidence-archive-history.json", "summary": { "aCaseCount": 680, "manifestedCases": 680, "missingOriginalEvidence": 0, "missingArtifact": 0, "snapshottedCases": 600, "partialSnapshotCases": 0, "snapshotAttentionCases": 1, "pendingSnapshotCases": 79, "snapshotIssueTypes": 2, "uniqueHosts": 137, "targets": 1365 }, "snapshotIssues": [ { "code": "pending_queued", "label": "未开始抓取", "description": "当前案例尚未生成本地 manifest，仍在按 host 分批抓取的队列中。", "action": "按 /updates 的 host 队列运行 npm run evidence:snapshot。", "cases": 79, "targets": 160, "highPriorityCases": 42, "hosts": [ { "host": "bilibili.com", "cases": 6, "targets": 12 }, { "host": "openai.com", "cases": 8, "targets": 8 }, { "host": "seed.bytedance.com", "cases": 8, "targets": 8 }, { "host": "openrouter.ai", "cases": 4, "targets": 8 }, { "host": "fal.ai", "cases": 3, "targets": 6 }, { "host": "youtube.com", "cases": 3, "targets": 5 }, { "host": "build.nvidia.com", "cases": 2, "targets": 4 }, { "host": "deepinfra.com", "cases": 2, "targets": 4 } ] }, { "code": "evidence_changed", "label": "证据字段变化", "description": "案例字段或证据 URL 与本地 manifest 指纹不一致，历史快照需要重新核对。", "action": "人工确认字段变化是否合理，再重跑该案例快照。", "cases": 1, "targets": 2, "highPriorityCases": 1, "hosts": [ { "host": "anthropic.com", "cases": 1, "targets": 1 }, { "host": "perplexity.ai", "cases": 1, "targets": 1 } ] } ], "latestHistory": { "id": "current-2026-07-27", "date": "2026-07-27", "label": "当前数据截点", "aCaseCount": 680, "targets": 1365, "snapshottedCases": 600, "pendingSnapshotCases": 79, "partialSnapshotCases": 0, "snapshotAttentionCases": 1, "source": "site/src/data/evidence-archive.json" } }, "featuredModels": [ { "id": "glm-4-7", "name": "GLM-4.7", "vendor": "Z AI / GLM", "url": "https://model-seven-blond.vercel.app/models/glm-4-7", "aCaseCount": 13, "score": 42.11, "publishability": "Publishable" }, { "id": "gpt-5-1-high", "name": "GPT-5.1 (high)", "vendor": "OpenAI", "url": "https://model-seven-blond.vercel.app/models/gpt-5-1-high", "aCaseCount": 12, "score": 47.7, "publishability": "Publishable" }, { "id": "qwen3-max", "name": "Qwen3 Max", "vendor": "Qwen / Alibaba", "url": "https://model-seven-blond.vercel.app/models/qwen3-max", "aCaseCount": 11, "score": 31.38, "publishability": "Publishable" }, { "id": "glm-5-1", "name": "GLM-5.1", "vendor": "Z AI / GLM", "url": "https://model-seven-blond.vercel.app/models/glm-5-1", "aCaseCount": 10, "score": 51.41, "publishability": "Publishable" }, { "id": "solar-pro-3", "name": "Solar Pro 3", "vendor": "Upstage / Solar", "url": "https://model-seven-blond.vercel.app/models/solar-pro-3", "aCaseCount": 10, "score": 25.87, "publishability": "Publishable" }, { "id": "grok-beta", "name": "Grok Beta", "vendor": "xAI / Grok", "url": "https://model-seven-blond.vercel.app/models/grok-beta", "aCaseCount": 10, "score": 13.28, "publishability": "Publishable" }, { "id": "gemini-1-5-pro-may", "name": "Gemini 1.5 Pro (May)", "vendor": "Google / Gemini", "url": "https://model-seven-blond.vercel.app/models/gemini-1-5-pro-may", "aCaseCount": 10, "score": 12, "publishability": "Publishable" }, { "id": "minimax-m1-80k", "name": "MiniMax M1 80k", "vendor": "MiniMax", "url": "https://model-seven-blond.vercel.app/models/minimax-m1-80k", "aCaseCount": 9, "score": 24.43, "publishability": "Publishable" }, { "id": "grok-2", "name": "Grok 2", "vendor": "xAI / Grok", "url": "https://model-seven-blond.vercel.app/models/grok-2", "aCaseCount": 9, "score": 13.89, "publishability": "Publishable" }, { "id": "solar-mini", "name": "Solar Mini", "vendor": "Upstage / Solar", "url": "https://model-seven-blond.vercel.app/models/solar-mini", "aCaseCount": 9, "score": 11.9, "publishability": "Publishable" }, { "id": "seedance-1-0", "name": "Seedance 1.0", "vendor": "ByteDance Seed", "url": "https://model-seven-blond.vercel.app/models/seedance-1-0", "aCaseCount": 9, "score": null, "publishability": "Publishable" }, { "id": "grok-4-3-high", "name": "Grok 4.3 (high)", "vendor": "xAI / Grok", "url": "https://model-seven-blond.vercel.app/models/grok-4-3-high", "aCaseCount": 8, "score": 53.2, "publishability": "Publishable" }, { "id": "mimo-v2-flash-feb-2026", "name": "MiMo-V2-Flash (Feb 2026)", "vendor": "Xiaomi / MiMo", "url": "https://model-seven-blond.vercel.app/models/mimo-v2-flash-feb-2026", "aCaseCount": 8, "score": 41.46, "publishability": "Publishable" }, { "id": "claude-fable-5", "name": "Claude Fable 5", "vendor": "Anthropic / Claude", "url": "https://model-seven-blond.vercel.app/models/claude-fable-5", "aCaseCount": 8, "score": null, "publishability": "Publishable" }, { "id": "seed1-5-vl", "name": "Seed1.5 VL", "vendor": "ByteDance Seed", "url": "https://model-seven-blond.vercel.app/models/seed1-5-vl", "aCaseCount": 8, "score": null, "publishability": "Publishable" }, { "id": "qwen3-6-max-preview", "name": "Qwen3.6 Max Preview", "vendor": "Qwen / Alibaba", "url": "https://model-seven-blond.vercel.app/models/qwen3-6-max-preview", "aCaseCount": 7, "score": 51.81, "publishability": "Publishable" }, { "id": "deepseek-v4-pro-max", "name": "DeepSeek V4 Pro (Max)", "vendor": "DeepSeek", "url": "https://model-seven-blond.vercel.app/models/deepseek-v4-pro-max", "aCaseCount": 7, "score": 51.51, "publishability": "Publishable" }, { "id": "deepseek-v3-1-terminus", "name": "DeepSeek V3.1 Terminus", "vendor": "DeepSeek", "url": "https://model-seven-blond.vercel.app/models/deepseek-v3-1-terminus", "aCaseCount": 7, "score": 33.93, "publishability": "Publishable" }, { "id": "o1", "name": "o1", "vendor": "OpenAI", "url": "https://model-seven-blond.vercel.app/models/o1", "aCaseCount": 7, "score": 30.75, "publishability": "Publishable" }, { "id": "o1-preview", "name": "o1-preview", "vendor": "OpenAI", "url": "https://model-seven-blond.vercel.app/models/o1-preview", "aCaseCount": 7, "score": 23.74, "publishability": "Publishable" } ], "recentCases": [ { "id": "gemini-1-5-pro-qodo-code-review", "title": "Qodo (formerly CodiumAI) 使用 Gemini 1.5 Pro (May) 处理代码审查和测试生成", "originalTitle": "Qodo (formerly CodiumAI) Integrates Gemini 1.5 Pro for Automated Code Review and Test Generation", "summary": "Qodo (formerly CodiumAI) 公开的代码审查与测试案例，来源为 qodo.ai，复核于 2026-06-27T19:00:00Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "代码审查与测试", "modelName": "Gemini 1.5 Pro (May)", "vendor": "Google / Gemini", "sourcePlatform": "qodo.ai", "collectedAt": "2026-06-27T19:00:00Z", "evidenceTrust": { "score": 99, "tier": "A+ 完整链路", "sourceKind": "官方/客户故事" }, "url": "https://model-seven-blond.vercel.app/cases/gemini-1-5-pro-qodo-code-review", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=gemini-1-5-pro-qodo-code-review" }, { "id": "grok-4-20-0309-v2-alphaone-ai-memory-a2a", "title": "AlphaOne LLC 使用 Grok 4.20 0309 v2 处理智能体流程编排", "originalTitle": "AlphaOne 的 ai-memory-a2a 用 Grok 4.20 0309 Reasoning 驱动记忆场景推理", "summary": "AlphaOne LLC 公开的智能体工作流案例，来源为 公开代码库，复核于 2026-06-27T10:42:00Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "智能体工作流", "modelName": "Grok 4.20 0309 v2", "vendor": "xAI / Grok", "sourcePlatform": "GitHub", "collectedAt": "2026-06-27T10:42:00Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/grok-4-20-0309-v2-alphaone-ai-memory-a2a", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=grok-4-20-0309-v2-alphaone-ai-memory-a2a" }, { "id": "grok-4-20-0309-v2-lumiwealth-lumibot-m2", "title": "Lumiwealth / Lumibot 使用 Grok 4.20 0309 v2 处理代码审查和测试生成", "originalTitle": "Lumiwealth 的 Lumibot 示例用 Grok 4.20 0309 Reasoning 做 M2 流动性交易策略分析", "summary": "Lumiwealth / Lumibot 公开的代码审查与测试案例，来源为 公开代码库，复核于 2026-06-27T10:42:00Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "代码审查与测试", "modelName": "Grok 4.20 0309 v2", "vendor": "xAI / Grok", "sourcePlatform": "GitHub", "collectedAt": "2026-06-27T10:42:00Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/grok-4-20-0309-v2-lumiwealth-lumibot-m2", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=grok-4-20-0309-v2-lumiwealth-lumibot-m2" }, { "id": "seed2-1-pro-volcengine-ark-api-experience-20260627", "title": "火山引擎方舟（Volcengine Ark） 使用 Seed2.1 Pro 处理智能体流程编排", "originalTitle": "火山方舟上线 Doubao-Seed-2.1-Pro，用于开发者 API 调用和体验中心试用", "summary": "火山引擎方舟（Volcengine Ark） 公开的智能体工作流案例，来源为 官方页面，复核于 2026-06-27T10:37:03Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "智能体工作流", "modelName": "Seed2.1 Pro", "vendor": "ByteDance Seed", "sourcePlatform": "official_web", "collectedAt": "2026-06-27T10:37:03Z", "evidenceTrust": { "score": 99, "tier": "A+ 完整链路", "sourceKind": "官方/客户故事" }, "url": "https://model-seven-blond.vercel.app/cases/seed2-1-pro-volcengine-ark-api-experience-20260627", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=seed2-1-pro-volcengine-ark-api-experience-20260627" }, { "id": "seed2-1-pro-volcengine-ark-experience-api-20260627", "title": "火山方舟（Volcengine Ark） 使用 Seed2.1 Pro 处理智能体流程编排", "originalTitle": "火山方舟上线 Doubao-Seed-2.1-Pro 体验中心与 API，用于开发者真实调用 Agent/Coding 模型能力", "summary": "火山方舟（Volcengine Ark） 公开的智能体工作流案例，来源为 官方页面，复核于 2026-06-27T10:34:04Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "智能体工作流", "modelName": "Seed2.1 Pro", "vendor": "ByteDance Seed", "sourcePlatform": "official_web", "collectedAt": "2026-06-27T10:34:04Z", "evidenceTrust": { "score": 99, "tier": "A+ 完整链路", "sourceKind": "官方/客户故事" }, "url": "https://model-seven-blond.vercel.app/cases/seed2-1-pro-volcengine-ark-experience-api-20260627", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=seed2-1-pro-volcengine-ark-experience-api-20260627" }, { "id": "claude-mythos-5-coasty-open-cowork", "title": "coasty-ai / Coasty AI 使用 Claude Mythos 5 处理智能体流程编排", "originalTitle": "Coasty AI open-cowork：用 Claude Mythos 5 构建开源 AI coworker 桌面/浏览器自动化项目", "summary": "coasty-ai / Coasty AI 公开的智能体工作流案例，来源为 公开代码库，复核于 2026-06-27T10:32:50Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "智能体工作流", "modelName": "Claude Mythos 5", "vendor": "Anthropic / Claude", "sourcePlatform": "github", "collectedAt": "2026-06-27T10:32:50Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/claude-mythos-5-coasty-open-cowork", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=claude-mythos-5-coasty-open-cowork" }, { "id": "claude-mythos-5-thibaultbm-claude-seo-geo", "title": "Thibaultbm / Sorank 使用 Claude Mythos 5 处理软件工程任务执行", "originalTitle": "Claude SEO GEO：用 Claude Mythos 5 构建面向 Claude Code 的 SEO/GEO 技能包", "summary": "Thibaultbm / Sorank 公开的代码代理与软件工程案例，来源为 公开代码库，复核于 2026-06-27T10:32:50Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "代码代理与软件工程", "modelName": "Claude Mythos 5", "vendor": "Anthropic / Claude", "sourcePlatform": "github", "collectedAt": "2026-06-27T10:32:50Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/claude-mythos-5-thibaultbm-claude-seo-geo", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=claude-mythos-5-thibaultbm-claude-seo-geo" }, { "id": "claude-mythos-5-shubhamsinghrathore-claude-skills-20260627", "title": "shubhamsinghrathore 使用 Claude Mythos 5 处理知识检索和问答", "originalTitle": "shubhamsinghrathore 用 Claude Mythos 5 创建 DevOps 调试类 Claude Skills", "summary": "shubhamsinghrathore 公开的知识库与检索问答案例，来源为 公开代码库，复核于 2026-06-27T10:32:15Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "知识库与检索问答", "modelName": "Claude Mythos 5", "vendor": "Anthropic / Claude", "sourcePlatform": "github", "collectedAt": "2026-06-27T10:32:15Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/claude-mythos-5-shubhamsinghrathore-claude-skills-20260627", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=claude-mythos-5-shubhamsinghrathore-claude-skills-20260627" }, { "id": "seed2-1-turbo-long-movies-narrated-shorts-20260627", "title": "ByteDance Seed 使用 Seed2.1 Turbo 处理多模态内容处理", "originalTitle": "ByteDance Seed showcases Seed2.1 Turbo for turning long movies into narrated shorts", "summary": "ByteDance Seed 公开的多模态生成与理解案例，来源为 官方页面，复核于 2026-06-27T10:31:20Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "多模态生成与理解", "modelName": "Seed2.1 Turbo", "vendor": "ByteDance Seed", "sourcePlatform": "official_web", "collectedAt": "2026-06-27T10:31:20Z", "evidenceTrust": { "score": 99, "tier": "A+ 完整链路", "sourceKind": "官方/客户故事" }, "url": "https://model-seven-blond.vercel.app/cases/seed2-1-turbo-long-movies-narrated-shorts-20260627", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=seed2-1-turbo-long-movies-narrated-shorts-20260627" }, { "id": "seed2-1-turbo-long-movie-narrated-shorts-showcase-20260627", "title": "ByteDance Seed 使用 Seed2.1 Turbo 处理多模态内容处理", "originalTitle": "ByteDance Seed showcases Seed2.1 Turbo for turning long movies into narrated short videos", "summary": "ByteDance Seed 公开的多模态生成与理解案例，来源为 官方页面，复核于 2026-06-27T10:29:26Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "多模态生成与理解", "modelName": "Seed2.1 Turbo", "vendor": "ByteDance Seed", "sourcePlatform": "official_web", "collectedAt": "2026-06-27T10:29:26Z", "evidenceTrust": { "score": 99, "tier": "A+ 完整链路", "sourceKind": "官方/客户故事" }, "url": "https://model-seven-blond.vercel.app/cases/seed2-1-turbo-long-movie-narrated-shorts-showcase-20260627", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=seed2-1-turbo-long-movie-narrated-shorts-showcase-20260627" }, { "id": "grok-4-20-0309-v2-swift-innovate-golem", "title": "swift-innovate 使用 Grok 4.20 0309 v2 处理代码审查和测试生成", "originalTitle": "swift-innovate 的 GOLEM 使用 Grok 4.20 0309 Reasoning 作为自改进 LangGraph Agent 主脑", "summary": "swift-innovate 公开的代码审查与测试案例，来源为 公开代码库，复核于 2026-06-27T10:28:54Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "代码审查与测试", "modelName": "Grok 4.20 0309 v2", "vendor": "xAI / Grok", "sourcePlatform": "GitHub", "collectedAt": "2026-06-27T10:28:54Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/grok-4-20-0309-v2-swift-innovate-golem", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=grok-4-20-0309-v2-swift-innovate-golem" }, { "id": "claude-2-0-librechat-multimodel-chat-20260627", "title": "LibreChat / danny-avila 使用 Claude 2.0 处理软件工程任务执行", "originalTitle": "LibreChat supported Claude 2 as an Anthropic model option for self-hosted multi-model chat", "summary": "LibreChat / danny-avila 公开的代码代理与软件工程案例，来源为 公开代码库，复核于 2026-06-27T10:28:36Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "代码代理与软件工程", "modelName": "Claude 2.0", "vendor": "Anthropic / Claude", "sourcePlatform": "github", "collectedAt": "2026-06-27T10:28:36Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/claude-2-0-librechat-multimodel-chat-20260627", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=claude-2-0-librechat-multimodel-chat-20260627" }, { "id": "grok-4-20-0309-bondartsov-grok-critic-mcp", "title": "Bondartsov 使用 Grok 4.20 0309 处理代码审查和测试生成", "originalTitle": "Bondartsov 的 grok-critic-mcp 用 Grok 4.20 Multi-Agent 做代码审查、架构分析和安全审计", "summary": "Bondartsov 公开的代码审查与测试案例，来源为 公开代码库，复核于 2026-06-27T10:26:24Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "代码审查与测试", "modelName": "Grok 4.20 0309", "vendor": "xAI / Grok", "sourcePlatform": "GitHub", "collectedAt": "2026-06-27T10:26:24Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/grok-4-20-0309-bondartsov-grok-critic-mcp", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=grok-4-20-0309-bondartsov-grok-critic-mcp" }, { "id": "gpt-5-1-high-virona-shawq-creative-chat-20260109", "title": "elmorshedy-del / Virona-ShawQ-Dashb… 使用 GPT-5.1 (high) 处理研究分析和报告生成", "originalTitle": "Virona ShawQ Dashboard adds GPT-5.1 high-effort reasoning to Creative Intelligence chat", "summary": "elmorshedy-del / Virona-ShawQ-Dashboard 公开的研究与报告生成案例，来源为 公开代码库，复核于 2026-06-27T10:25:47Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "研究与报告生成", "modelName": "GPT-5.1 (high)", "vendor": "OpenAI", "sourcePlatform": "github", "collectedAt": "2026-06-27T10:25:47Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/gpt-5-1-high-virona-shawq-creative-chat-20260109", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=gpt-5-1-high-virona-shawq-creative-chat-20260109" }, { "id": "gpt-5-1-high-overhypeme-ai-prompts-20260527", "title": "TheAnswerManIsHere / Overhypeme 使用 GPT-5.1 (high) 处理多模态内容处理", "originalTitle": "Overhypeme enables GPT-5.1 high-effort reasoning for AI image style and video motion prompts", "summary": "TheAnswerManIsHere / Overhypeme 公开的多模态生成与理解案例，来源为 公开代码库，复核于 2026-06-27T10:25:47Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "多模态生成与理解", "modelName": "GPT-5.1 (high)", "vendor": "OpenAI", "sourcePlatform": "github", "collectedAt": "2026-06-27T10:25:47Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/gpt-5-1-high-overhypeme-ai-prompts-20260527", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=gpt-5-1-high-overhypeme-ai-prompts-20260527" }, { "id": "gpt-5-1-high-latitude-azure-prompt-routing-20260515", "title": "latitude-dev / latitude-llm 使用 GPT-5.1 (high) 处理知识检索和问答", "originalTitle": "Latitude fixes Azure GPT-5.1 high-effort prompt routing for its LLM platform", "summary": "latitude-dev / latitude-llm 公开的知识库与检索问答案例，来源为 公开代码库，复核于 2026-06-27T10:25:47Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "知识库与检索问答", "modelName": "GPT-5.1 (high)", "vendor": "OpenAI", "sourcePlatform": "github", "collectedAt": "2026-06-27T10:25:47Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/gpt-5-1-high-latitude-azure-prompt-routing-20260515", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=gpt-5-1-high-latitude-azure-prompt-routing-20260515" }, { "id": "gpt-5-1-high-vscode-azure-byok-20260523", "title": "Microsoft / Visual Studio Code 使用 GPT-5.1 (high) 处理智能体流程编排", "originalTitle": "VS Code forwards Azure BYOK GPT-5.1 high reasoning effort through the Responses API", "summary": "Microsoft / Visual Studio Code 公开的智能体工作流案例，来源为 公开代码库，复核于 2026-06-27T10:25:47Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "智能体工作流", "modelName": "GPT-5.1 (high)", "vendor": "OpenAI", "sourcePlatform": "github", "collectedAt": "2026-06-27T10:25:47Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/gpt-5-1-high-vscode-azure-byok-20260523", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=gpt-5-1-high-vscode-azure-byok-20260523" }, { "id": "gpt-5-1-high-iii-workers-openai-provider-20260612", "title": "iii-hq / workers 使用 GPT-5.1 (high) 处理真实任务执行", "originalTitle": "iii-hq Workers adds an OpenAI provider worker with GPT-5.1 high-effort reasoning support", "summary": "iii-hq / workers 公开的真实任务执行案例，来源为 公开代码库，复核于 2026-06-27T10:25:47Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "真实任务执行", "modelName": "GPT-5.1 (high)", "vendor": "OpenAI", "sourcePlatform": "github", "collectedAt": "2026-06-27T10:25:47Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/gpt-5-1-high-iii-workers-openai-provider-20260612", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=gpt-5-1-high-iii-workers-openai-provider-20260612" }, { "id": "solar-pro-2-ho323-korean-modernization", "title": "ho323 / MixUP Upstage Promptathon p… 使用 Solar Pro 2 处理软件工程任务执行", "originalTitle": "MixUP Upstage project uses Solar Pro2 for Korean historical-text modernization", "summary": "ho323 / MixUP Upstage Promptathon participant 公开的代码代理与软件工程案例，来源为 公开代码库，复核于 2026-06-27T10:25:34Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "代码代理与软件工程", "modelName": "Solar Pro 2", "vendor": "Upstage / Solar", "sourcePlatform": "GitHub", "collectedAt": "2026-06-27T10:25:34Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/solar-pro-2-ho323-korean-modernization", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=solar-pro-2-ho323-korean-modernization" }, { "id": "solar-pro-2-geminii01-upthink", "title": "geminii01 / UpThink project team 使用 Solar Pro 2 处理软件工程任务执行", "originalTitle": "UpThink uses Solar Pro 2 to automate Obsidian personal knowledge management", "summary": "geminii01 / UpThink project team 公开的代码代理与软件工程案例，来源为 公开代码库，复核于 2026-06-27T10:25:34Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。", "useCase": "代码代理与软件工程", "modelName": "Solar Pro 2", "vendor": "Upstage / Solar", "sourcePlatform": "GitHub", "collectedAt": "2026-06-27T10:25:34Z", "evidenceTrust": { "score": 100, "tier": "A+ 完整链路", "sourceKind": "代码仓库证据" }, "url": "https://model-seven-blond.vercel.app/cases/solar-pro-2-geminii01-upthink", "evidenceDrawerUrl": "https://model-seven-blond.vercel.app/cases?case=solar-pro-2-geminii01-upthink" } ] }