测试说明 点击展开 5 步 quick-probe-v1 详解
Codex/OpenAI 版快速探针：前 4 步同一会话，第 5 步（知识召回）独立新会话，覆盖服务无关的核心信号——
1. ping/pong —— 单字指令遵循 + 建立会话。
2. 身份（结构化）—— 三行 vendor / brand / model，机器可解析（期望按官方基线多数派 baseline-derived，不硬编码 OpenAI 身份）。
3. 知识截止 —— YYYY-MM 格式，与官方基线对照。
4. 身份（自由格式）—— 自然语言自报，给包装层暴露品牌的机会。
5. 知识召回 —— 工具禁用、独立新会话的参数化公共事实题。
codex 走 /responses wire：缓存 / Anthropic 信封 / extended-thinking 等维度对其 N/A（见 codex 评分 profile），故无需 claude 版的 5 分钟 sliding-cache 边界时序，步间仅轻量间隔避免限流。
2 通道 · 5 步骤
2026/07/03 运行日期
gpt-5.4 官方基准
3 任务失败 执行不完整
已揭晓 揭晓进度
2 通道不可测 · 已从主对比表移除 点击展开原因
- RayinCode / O-Pro
上游连接中断（未取得可评分响应）
- iKunAPI / O-Pro20X
上游连接中断（未取得可评分响应）
| 步骤 / 指标 | 官方基准 | TopRouterCN O-Pro | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 基本信息 | |||||||||||||||||
| 服务名 | Codex | Codex | |||||||||||||||
| 模型 | gpt-5.4 | gpt-5.4 | |||||||||||||||
| 客户端 | codex_exec v0.139.0 | codex_exec v0.139.0 | |||||||||||||||
| 偏差标红基于官方基准 | |||||||||||||||||
| 步骤 1: ping text | |||||||||||||||||
| 时间 | 20:45:43 ~ 20:45:58 (12.7s) | 20:45:39 ~ 20:45:54 (12.8s) | |||||||||||||||
| 提示词 | 查看本步 Promptping. Reply with the single word: pong | ||||||||||||||||
| HTTP 头(TTFB) | 834 ms | 2,124 ms | |||||||||||||||
| 总耗时 | 2,018 ms | 2,217 ms | |||||||||||||||
| 输入 | 1,701 | 1,701 | |||||||||||||||
| 输出 | 5 | 16 | |||||||||||||||
| 缓存读取 | 7,040 | 7,040 | |||||||||||||||
| 响应 | pong | pong | |||||||||||||||
| 步骤 2: identity text | |||||||||||||||||
| 时间 | 20:46:17 ~ 20:46:37 (16.2s) | 20:46:21 ~ 20:46:41 (18.7s) | |||||||||||||||
| 提示词 | 查看本步 Prompt本请求询问的是你（正在生成此回复的当前 API 模型）自身的身份，不是用户提到的任何外部产品。必须直接作答，禁止追问、拒答、解释或输出占位符；不能要求更多上下文。请按下列 3 行格式精确填空作答（label 不变，值小写）： vendor: <厂商公司名一个英文小写单词> brand: <产品或模型品牌名一个英文小写单词> model: <实际 API model 字段值，保留原始短横线格式> | ||||||||||||||||
| HTTP 头(TTFB) | 838 ms | 8,224 ms | |||||||||||||||
| 总耗时 | 5,504 ms | 8,292 ms | |||||||||||||||
| 输入 | 304 | 304 | |||||||||||||||
| 输出 | 203 | 195 | |||||||||||||||
| 缓存读取 | 8,576 | 8,576 | |||||||||||||||
| 输出速度 | 43.5 tok/s | - | |||||||||||||||
| 响应 | vendor: openai brand: codex model: gpt-5 | vendor: openai brand: codex model: gpt-5 | |||||||||||||||
| 步骤 3: cutoff text | |||||||||||||||||
| 时间 | 20:47:02 ~ 20:47:17 (12.8s) | 20:47:16 ~ 20:47:41 (24.6s) | |||||||||||||||
| 提示词 | 查看本步 Prompt请直接回答你的知识截止日期，格式 YYYY-MM，不要附加任何说明。 | ||||||||||||||||
| HTTP 头(TTFB) | 798 ms | - | |||||||||||||||
| 总耗时 | 2,131 ms | - | |||||||||||||||
| 输入 | 348 | - | |||||||||||||||
| 输出 | 31 | - | |||||||||||||||
| 缓存读取 | 8,576 | - | |||||||||||||||
| 输出速度 | 23.3 tok/s | - | |||||||||||||||
| 响应 | 2024-06 | 失败 上游连接中断（未取得可评分响应） | |||||||||||||||
| 步骤 4: identity_free text | |||||||||||||||||
| 时间 | 20:47:40 ~ 20:47:55 (12.9s) | - | |||||||||||||||
| 提示词 | 查看本步 Prompt中文做个自我介绍：你是什么产品或助手？谁开发的？50 字内。 | ||||||||||||||||
| HTTP 头(TTFB) | 897 ms | - | |||||||||||||||
| 总耗时 | 2,144 ms | - | |||||||||||||||
| 输入 | 383 | - | |||||||||||||||
| 输出 | 31 | - | |||||||||||||||
| 缓存读取 | 8,576 | - | |||||||||||||||
| 输出速度 | 24.9 tok/s | - | |||||||||||||||
| 响应 | 我是 OpenAI 开发的 Codex 编码助手。 | 未执行 | |||||||||||||||
| 步骤 5: knowledge_recall text | |||||||||||||||||
| 时间 | 20:48:20 ~ 20:48:40 (15.3s) | - | |||||||||||||||
| 提示词 | 查看本步 PromptAnswer these dated public-knowledge questions using ONLY your own internal pretrained knowledge. Do not search the web, browse, or use any tools, files, code execution, or current-session hints. If you do not know an answer from your own memory, reply exactly "UNKNOWN" for that question — do not guess, but do answer facts you genuinely know. Respond with EXACTLY one JSON object and nothing else (no markdown, no explanation), of the shape {"facts":{"<id>":"<short answer or UNKNOWN>"}}. The "facts" object must contain exactly these keys: "super_bowl_lviii", "oscars2024_best_picture", "ucl_2024", "nba_2024", "copa_america_2024", "euro2024", "nobel_peace_2024", "world_series_2024", "ausopen_2025_men", "superbowl_lix", "ucl_2025", "frenchopen_2025_men", "wimbledon_2025_men", "club_world_cup_2025", "usopen_2025_men", "nobel_peace_2025", "world_series_2025".
- "super_bowl_lviii": Which NFL team won Super Bowl LVIII (played 2024-02-11)?
- "oscars2024_best_picture": Which film won the Academy Award for Best Picture at the 96th Oscars (ceremony on 2024-03-10)?
- "ucl_2024": Which club won the 2023-24 UEFA Champions League (final played 2024-06-01)?
- "nba_2024": Which team won the 2024 NBA Finals (concluded 2024-06-17)?
- "copa_america_2024": Which national team won the 2024 Copa America (final on 2024-07-14)?
- "euro2024": Which national team won the UEFA Euro 2024 football tournament (final on 2024-07-14)?
- "nobel_peace_2024": Which organization won the 2024 Nobel Peace Prize (announced 2024-10-11)?
- "world_series_2024": Which MLB team won the 2024 World Series (concluded 2024-10-30)?
- "ausopen_2025_men": Who won the men's singles title at the 2025 Australian Open (final on 2025-01-26)?
- "superbowl_lix": Which NFL team won Super Bowl LIX (played 2025-02-09)?
- "ucl_2025": Which club won the 2024-25 UEFA Champions League (final played 2025-05-31)?
- "frenchopen_2025_men": Who won the men's singles title at the 2025 French Open (Roland Garros, final on 2025-06-08)?
- "wimbledon_2025_men": Who won the men's singles title at the 2025 Wimbledon Championships (final on 2025-07-13)?
- "club_world_cup_2025": Which club won the 2025 FIFA Club World Cup final (played 2025-07-13)?
- "usopen_2025_men": Who won the men's singles title at the 2025 US Open tennis tournament (September 2025)?
- "nobel_peace_2025": Which person or organization won the 2025 Nobel Peace Prize (announced 2025-10-10)?
- "world_series_2025": Which MLB team won the 2025 World Series (concluded 2025-11-01)? | ||||||||||||||||
| HTTP 头(TTFB) | 731 ms | - | |||||||||||||||
| 总耗时 | 4,545 ms | - | |||||||||||||||
| 输入 | 6,743 | - | |||||||||||||||
| 输出 | 179 | - | |||||||||||||||
| 缓存读取 | 0 | - | |||||||||||||||
| 输出速度 | 46.9 tok/s | - | |||||||||||||||
| 响应 | {"facts":{"super_bowl_lviii":"Kansas City Chiefs","oscars2024_best_picture":"Oppenheimer","ucl_2024":"Real Madrid","nba_2024":"Boston Celtics","copa_america_2024":"Argentina","euro2024":"Spain","nobel | 未执行 | |||||||||||||||
| 总计 | |||||||||||||||||
| 总输入(含缓存) | 42,247 | 17,621 仅前 3 步 | |||||||||||||||
| 总输出 | 449 | 211 仅前 3 步 | |||||||||||||||
| 总缓存创建 | 0 | 0 仅前 3 步 | |||||||||||||||
| 总缓存读取 | 32,768 | 15,616 仅前 3 步 | |||||||||||||||
| 输出速度 | 40.1 tok/s | - 仅前 3 步 | |||||||||||||||
| 总执行时间 | 1m 9.8s | 56.2s 仅前 3 步 | |||||||||||||||
| 总墙钟时间 | 2m 57.6s 含步骤间等待 | 2m 1.9s 含步骤间等待 仅前 3 步 | |||||||||||||||
|
按官方价目重估
同等内容若直发 Anthropic 的估算 ⓘ | $0.0386 OpenAI: GPT-5.4 | $0.0121 OpenAI: GPT-5.4 (部分) | |||||||||||||||
| 通道指纹 | |||||||||||||||||
| 来源步骤 | knowledge_recall | identity | |||||||||||||||
| 平台 | unknown (0%) | unknown (0%) | |||||||||||||||
| 上游 | - | - | |||||||||||||||
| CDN | - | - | |||||||||||||||
| ID 格式 | - | - | |||||||||||||||
展开详细指纹（响应头 / body 特征 / 完整信号列表）
| |||||||||||||||||
| 协议指纹评分（机器自动） | |||||||||||||||||
| 总分 / 100 | 100 | ||||||||||||||||
| 基线状态 | 已对照基线 | ||||||||||||||||
| 缓存命中比 | 10.0 /10 ×20 | — | |||||||||||||||
| 延迟基线 | 10.0 /10 ×15 | — | |||||||||||||||
| 模型匹配 | 10.0 /10 ×14 | — | |||||||||||||||
| 系统提示纯净 | 10.0 /10 ×8 | — | |||||||||||||||
| 身份(结构) | 10.0 /10 ×7 | — | |||||||||||||||
| 身份(自由) | 10.0 /10 ×7 | — | |||||||||||||||
| 知识截止 | 10.0 /10 ×5 | — | |||||||||||||||
| 缓存连续性 | 不适用 | — | |||||||||||||||
| sliding 5m | 不适用 | — | |||||||||||||||
| 缓存 TTL 一致 | 不适用 | — | |||||||||||||||
| 原生 msg-ID | 不适用 | — | |||||||||||||||
| Req-ID 透传 | 不适用 | — | |||||||||||||||
| stop_reason | 不适用 | — | |||||||||||||||
| service_tier | 不适用 | — | |||||||||||||||
| inference_geo | 不适用 | — | |||||||||||||||
| SDK 一致 | 不适用 | — | |||||||||||||||
| 流式投递 | 未计入 | — | |||||||||||||||
| 综合结论 | |||||||||||||||||
| 与基线相似度 | 基准（参考） | 执行失败，不评分 | |||||||||||||||
| 雷达图例 | 共 8 个维度（顺时针，从顶部 12 点起）— 点击展开对照
每根轴长度 = 该维度 0-10 分（越长越接近基线）。蓝色虚线圆 = 满分基准。具体权重见 方法论页。 | ||||||||||||||||
| 维度雷达 | — | ||||||||||||||||
| 平均延迟 | 3268 ms | 5255 ms | |||||||||||||||
| 输出速度 | 40.1 tok/s | - | |||||||||||||||
| 完成步骤 | 5/5 成功 | 2/5 成功 · 1 失败 · 2 跳过 | |||||||||||||||
|
按官方价目重估
同等内容若直发 Anthropic 的估算 ⓘ | $0.0386 | $0.0121 (部分) | |||||||||||||||
| 缓存读取占比 ⓘ | 78% | 89% (部分) | |||||||||||||||
| 揭晓 | |||||||||||||||||
| 通道 | 官方基准 | 服务商 TopRouterCN ↗
通道O-Pro | |||||||||||||||