可进入对比
档案摘要 / Decision snapshot
先判断能不能用于选型。
有真实任务证据
来自模型卡
边界优先
30 秒结论
可进入选型对比,但仍以案例证据为准。
来自公开模型资料库,并已补厂商/官方/模型家族证据;当前站内暂无可核验 A 类案例。
这页有来源、有风险说明、有公开可核验 A 类案例。
数据可信度 / 证据强度
强
A 类案例
5
AA 评分
53.56
官方来源
2 个补充入口
案例来源
5 条可核验
来自公开资料与模型数据库 + Artificial Analysis + 2 个厂商/官方/模型家族证据入口;AA 不是唯一事实来源。
Evidence Distribution
案例证据分布
GPT-5.3 Codex (xhigh) 当前关联 5 条 A 类案例;这里按任务、来源和复核时间观察证据结构。
Evidence Snapshot
证据快照覆盖
GPT-5.3 Codex (xhigh) 的 5 条 A 类案例中,已有 5 条完成本地证据快照;0 条仍在快照队列。
可用 manifest 复核
等待落盘
需补抓
需人工复核
证据完整度
5 / 5
Feishu Bitable
2 个入口
5 条
已绑定
已标注
能力边界
用标签表达信号,不用标签替代证据。
通用/待核验
暂无数据
待核验
平台/待核验
中
| 厂商 | OpenAI |
|---|---|
| 厂商页 | OpenAI |
| 发布时间 | 2026-02-05 |
| 模型 ID | gpt-5-3-codex-xhigh |
| 上下文 | 暂无数据 |
| 输出 | 暂无数据 |
| 模态 | 文本 / 多模态待核验 |
| 推理 | 是 / 推理模型或 thinking 模式 |
| 价格 | 官方未披露 / 暂无数据 |
| 可用平台 | OpenAI |
| AA 评分 | 53.56 |
| 基础模型发布时间 | Aug 31, 2025 |
| 资料来源 | 公开资料、厂商信息和案例库 |
| A 类案例 | 5 |
谱系位置
它在厂商路线中的位置
此页将 GPT-5.3 Codex (xhigh) 放入 OpenAI 的当前 Atlas 路线中。若同厂前后代资料不足,先保留为路线追踪入口,不脑补谱系关系。
2026-02-05
GPT-5.3 Codex (xhigh) 资料状态
来自公开资料与模型数据库 + Artificial Analysis + 2 个厂商/官方/模型家族证据入口;AA 不是唯一事实来源。
公开档案
模型状态
GPT-5.3 Codex (xhigh) 当前标记为已有真实案例。
案例库
公开案例复核
已有 5 条可核验 A 类案例,达到完整补齐线
案例时间线
最近进入档案的 A 类案例
用采集时间呈现证据进入 Atlas 的顺序,方便复核来源新鲜度。
2026-06-27T10:04:54Z
panggi used GPT-5.3 Codex xhigh to generate Zircon Common Lisp books
panggi / technical_writing_code_documentation
2026-06-27T10:04:54Z
Kenpal-ElenaTokumura generated a Next.js Minesweeper app with GPT-5.3 Codex in Copilot Agent mode
Kenpal-ElenaTokumura / software_development_game
2026-06-27T10:04:54Z
andreifie built the FLYPER VIBE educational game largely with GitHub Copilot and GPT-5.3 Codex
andreifie / software_development_education_game
2026-06-27T10:04:54Z
x1colegal vibe-coded the P2PHTTP peer-to-peer HTTP transport with GPT-5.3 Codex
x1colegal / software_development_networking
2026-06-27T10:04:54Z
minhnh created a Neovim plugin for robbdd/TextX languages using GPT-5.3 Codex
minhnh / software_development_developer_tooling
适合场景
- - 作为完整模型索引和厂商路线追踪入口
- - 推理/agentic workflow 候选
不适合
- - 缺少真实案例时不要包装成推荐
- - 缺字段时不要脑补价格、性能或上下文
真实案例
可核验 A 类案例精选
当前展示 5 条代表案例;排序优先看模型卡精选、证据可信度、快照状态和公开仓库/技术任务信号。案例总数仍为 5。
Representative Evidence
代表案例排序
当前第一代表案例是“andreifie built the FLYPER VIBE educational game largely with GitHub Copilot and GPT-5.3 Codex”。它会优先出现在本页,是因为该案例同时进入模型卡精选、具备 100/100 的证据可信度,并带有 已快照证据。
模型卡精选优先
A+ 完整链路
archive/evidence/gpt-5-3-codex-xhigh-andreifie-flypervibe/manifest.json
Coding / repo / agent
andreifie 使用 GPT-5.3 Codex (xhigh) 处理可玩交互原型构建
andreifie · GPT-5.3 Codex (xhigh)
A+ 完整链路 · 代码仓库证据
andreifie 公开的游戏与交互原型案例,来源为 公开代码库,复核于 2026-06-27T10:04:54Z。 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。
任务:这是一个围绕可玩交互原型构建的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Create an educational Phaser 3 game for engineering students to practice English grammar and vocabulary, with a Node.js/Express API, SQLite persistence, WebSocket social features, ranked play, lessons, friend challenges…
公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:A public full-stack game repository with client, server, database, real-time social features, API documentation, and local development/build commands.
模型作用:GPT-5.3 Codex (xhigh) 在该案例中承担可玩交互原型构建相关的生成、分析、编排或实现角色。 原始资料写作:The repository description says the project was almost entirely vibecoded with GitHub Copilot plus gpt-5.3-codex, tying the model to implementation of the game and backend features.
A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。
风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:The evidence names GPT-5.3-Codex without the xhigh suffix; included as a family-bound case for the GPT-5.3 Codex task model rather than as proof of a separately exposed xhigh runtime.
原始记录:andreifie built the FLYPER VIBE educational game largely with GitHub Copilot and GPT-5.3 Codex
Kenpal-ElenaTokumura 使用 GPT-5.3 Codex (xhigh) 处理可玩交互原型构建
Kenpal-ElenaTokumura · GPT-5.3 Codex (xhigh)
A+ 完整链路 · 代码仓库证据
Kenpal-ElenaTokumura 公开的游戏与交互原型案例,来源为 公开代码库,复核于 2026-06-27T10:04:54Z。 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。
任务:这是一个围绕可玩交互原型构建的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Build a Minesweeper game using Next.js, Bun, Tailwind CSS, lint/format/build scripts, and a documented local development workflow.
公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:A public repository for a working Next.js/Bun Minesweeper project with README instructions for installation, development, linting, formatting, building, and starting the app.
模型作用:GPT-5.3 Codex (xhigh) 在该案例中承担可玩交互原型构建相关的生成、分析、编排或实现角色。 原始资料写作:The repository metadata states it was generated by GPT-5.3-Codex in GitHub Copilot Agent mode, indicating the model produced the application code and project scaffolding.
A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。
风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:The public evidence names GPT-5.3-Codex but does not explicitly include the xhigh label; included because the task model is the GPT-5.3 Codex model entry and public model naming may omit reasoning-effort suffixes.
原始记录:Kenpal-ElenaTokumura generated a Next.js Minesweeper app with GPT-5.3 Codex in Copilot Agent mode
minhnh 使用 GPT-5.3 Codex (xhigh) 处理软件工程任务执行
minhnh · GPT-5.3 Codex (xhigh)
A+ 完整链路 · 代码仓库证据
minhnh 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 2026-06-27T10:04:54Z。 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。
任务:这是一个围绕软件工程任务执行的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Create a Neovim plugin for robbdd/TextX languages, including filetype detection, syntax highlighting fallback, optional LSP diagnostics and hover documentation via a bundled Python language server, setup configuration, …
公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:A public Neovim plugin repository with Lua/Python plugin code, README setup instructions, configuration examples, and Python/Neovim smoke test commands.
模型作用:GPT-5.3 Codex (xhigh) 在该案例中承担软件工程任务执行相关的生成、分析、编排或实现角色。 原始资料写作:The repository metadata states the Neovim plugin was created using GPT 5.3-Codex, tying the model to the generated developer-tooling implementation.
A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。
风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:The evidence names GPT 5.3-Codex without the xhigh suffix; included as a model-family-bound artifact because public GitHub descriptions commonly omit the reasoning-effort variant.
原始记录:minhnh created a Neovim plugin for robbdd/TextX languages using GPT-5.3 Codex
panggi 使用 GPT-5.3 Codex (xhigh) 处理软件工程任务执行
panggi · GPT-5.3 Codex (xhigh)
A+ 完整链路 · 代码仓库证据
panggi 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 2026-06-27T10:04:54Z。 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。
任务:这是一个围绕软件工程任务执行的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Generate book-style documentation/materials for Zircon Common Lisp, comparing outputs from Claude Code, Codex, and Kimi Code.
公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:A public GitHub repository containing generated Zircon Common Lisp book artifacts and a prompt file; the repository description explicitly says the Codex output used gpt-5.3-codex xhigh.
模型作用:GPT-5.3 Codex (xhigh) 在该案例中承担软件工程任务执行相关的生成、分析、编排或实现角色。 原始资料写作:GPT-5.3 Codex (xhigh) was one of the generation agents used to produce the Zircon Common Lisp book content, with the exact model variant named in the repository evidence.
A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。
风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:Evidence is a public GitHub repository description and artifact repo; content should still be reviewed if model-card use requires separating Codex output from other model outputs.
原始记录:panggi used GPT-5.3 Codex xhigh to generate Zircon Common Lisp books
x1colegal 使用 GPT-5.3 Codex (xhigh) 处理软件工程任务执行
x1colegal · GPT-5.3 Codex (xhigh)
A+ 完整链路 · 代码仓库证据
x1colegal 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 2026-06-27T10:04:54Z。 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。
任务:这是一个围绕软件工程任务执行的真实任务,公开材料可以回溯到具体使用者和具体产物。 原始资料写作:Prototype a peer-to-peer transport over plain HTTP/1.1 with peers serving object chunks, tracker-based peer discovery, HLS publish/ingest tooling, ffplay compatibility proxy, and a native Python/GStreamer player path.
公开产物:公开材料提供公开代码、README 或项目配置,可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作:A public Python repository implementing P2PHTTP, demo scripts for multi-peer operation, tracker/peer/client architecture documentation, and optional video playback tooling.
模型作用:GPT-5.3 Codex (xhigh) 在该案例中承担软件工程任务执行相关的生成、分析、编排或实现角色。 原始资料写作:The repository metadata explicitly says the project was vibe-coded with GPT-5.3-Codex, indicating the model contributed to generating the implementation and documentation for the networking prototype.
A 类理由:有具体使用者、具体任务、公开原始证据和可访问产物,可公开核验。
风险边界:当前判断基于公开材料;若产物下线、仓库变更或模型参与比例仅来自作者自述,需要在引用前重新复核。 原始资料写作:The evidence names GPT-5.3-Codex without the xhigh suffix; included as a public artifact-bound use case for the GPT-5.3 Codex model family.
原始记录:x1colegal vibe-coded the P2PHTTP peer-to-peer HTTP transport with GPT-5.3 Codex
数据缺口
只基于现有字段判断缺口;缺失项不会被猜测填充。
- - 补充方向:核验 价格 的官方/API/案例来源。
- - 补充方向:核验 上下文 的官方/API/案例来源。
- - 补充方向:核验 输出 的官方/API/案例来源。
下一步验证建议
- - 继续保留 A 类案例的原始证据、产物页和版本快照。
- - 补官方发布/API/System Card 来源;若官方未披露,继续标注“官方未披露”。
- - 把 benchmark、教程、发布文保留为背景资料,不提升为真实案例。