模型对比 | RPDiag

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。

5 通道 · 5 步骤

2026/07/02 运行日期

gpt-5.5 官方基准

2 通道分数 < 70 与基线相似度

已揭晓揭晓进度

步骤 / 指标

WanmoAPI

O-Pro

TopRouterCN

O-Pro

官方基准

AnyAIGC

M-Pro

ShareAi

基本信息

服务名 Codex Codex Codex Codex Codex

模型 gpt-5.5 gpt-5.5 gpt-5.5 gpt-5.5 gpt-5.5

客户端 codex_exec v0.139.0 codex_exec v0.139.0 codex_exec v0.139.0 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 15:10:04 ~ 15:10:19 (14.5s) 15:10:00 ~ 15:10:15 (12.7s) 15:10:08 ~ 15:10:23 (12.6s) 15:10:06 ~ 15:10:26 (16.0s) 15:10:02 ~ 15:10:17 (12.3s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 3,245 ms 2,168 ms 935 ms 4,465 ms 886 ms

总耗时 3,953 ms 2,276 ms 1,943 ms 5,051 ms 1,748 ms

输入 1,548 1,548 1,548 1,548 1,791

输出 5 5 5 5 5

缓存读取 8,576 8,576 8,576 8,576 3,968

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 15:10:45 ~ 15:11:10 (20.5s) 15:10:34 ~ 15:10:54 (19.4s) 15:10:43 ~ 15:10:58 (14.5s) 15:10:49 ~ 15:11:15 (24.6s) 15:10:47 ~ 15:11:07 (15.4s)

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。请只输出以下 3 行内容（label 小写，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) 6,998 ms 8,873 ms 992 ms 10,711 ms 797 ms

总耗时 9,954 ms 8,979 ms 3,767 ms 13,505 ms 4,809 ms

输入 145 1,681 145 1,681 3,972

输出 145 204 107 139 170

缓存读取 10,112 8,576 10,112 8,576 1,920

输出速度 49.1 tok/s - 38.6 tok/s 49.7 tok/s 42.4 tok/s

响应

vendor: openai
brand: codex
model: gpt-5-codex

vendor: openai
brand: codex
model: gpt-5-codex

vendor: openai
brand: codex
model: gpt-5

vendor: openai
brand: codex
model: gpt-5

vendor: openai
brand: codex
model: gpt-5-codex

步骤 3: cutoff text

时间 15:11:28 ~ 15:11:48 (15.8s) 15:11:30 ~ 15:11:45 (12.6s) 15:11:40 ~ 15:11:55 (12.5s) 15:11:42 ~ 15:12:02 (18.9s) 15:11:26 ~ 15:11:41 (13.3s)

提示词

查看本步 Prompt

你掌握的世界知识截止于何时？YYYY-MM 格式作答，不解释。

HTTP 头(TTFB) 4,362 ms 2,022 ms 991 ms 4,564 ms 855 ms

总耗时 5,214 ms 2,126 ms 1,808 ms 5,758 ms 2,820 ms

输入 194 1,720 1,727 1,725 437

输出 8 8 8 8 8

缓存读取 10,112 8,576 8,576 8,576 5,504

响应

2024-06

2024-06

2024-06

2024-06

2024-06

步骤 4: identity_free text

时间 15:12:05 ~ 15:12:25 (16.3s) 15:12:16 ~ 15:12:31 (13.8s) 15:12:23 ~ 15:12:38 (12.9s) 15:12:40 ~ 15:13:00 (16.2s) 15:12:10 ~ 15:12:25 (14.4s)

提示词

查看本步 Prompt

用中文回答：你是谁？由谁开发？50 字以内。

HTTP 头(TTFB) 4,163 ms 3,277 ms 903 ms 3,790 ms 821 ms

总耗时 5,767 ms 3,407 ms 2,251 ms 5,252 ms 3,933 ms

输入 1,759 216 223 221 466

输出 32 40 33 37 29

缓存读取 8,576 10,112 10,112 10,112 5,504

输出速度 20.0 tok/s - 24.5 tok/s 25.3 tok/s -

响应

我是 OpenAI 开发的 AI 助手，基于 GPT-5，可协助回答问题、写作与编程。

我是 OpenAI 开发的 Codex 编程助手，基于 GPT-5，擅长阅读代码、修改项目、运行命令并协助调试。

我是 OpenAI 打造的 Codex，基于 GPT-5 的编程助手，用于协助代码理解、修改与验证。

我是 OpenAI 开发的 Codex 编程助手，基于 GPT-5，擅长协助阅读、修改、调试和解释代码。

我是 OpenAI 的 Codex，由 OpenAI 开发，用于编程协助、代码分析与自动化任务。

步骤 5: knowledge_recall text

时间 15:12:44 ~ 15:13:04 (17.4s) 15:13:14 ~ 15:13:39 (21.7s) 15:13:19 ~ 15:13:39 (15.8s) 15:13:27 ~ 15:13:52 (22.2s) 15:12:42 ~ 15:13:02 (15.0s)

提示词

查看本步 Prompt

Answer 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) 4,601 ms 11,134 ms 1,018 ms 4,285 ms 809 ms

总耗时 6,912 ms 11,238 ms 5,124 ms 11,153 ms 4,610 ms

输入 8,126 8,126 6,590 8,157 3,761

输出 179 179 179 174 182

缓存读取 0 0 1,536 0 0

输出速度 77.5 tok/s - 43.6 tok/s 25.3 tok/s 47.9 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

{"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

{"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

{"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

{"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

总计

总输入(含缓存) 49,148 49,131 49,145 49,172 27,323

总输出 369 436 332 363 394

总缓存创建 0 0 0 0 0

总缓存读取 37,376 35,840 38,912 35,840 16,896

输出速度 51.8 tok/s - 38.8 tok/s 31.5 tok/s 45.1 tok/s

总执行时间 1m 24.5s 1m 20.2s 1m 8.3s 1m 37.8s 1m 10.4s

总墙钟时间

2m 59.8s

含步骤间等待

3m 38.9s

含步骤间等待

3m 30.9s

含步骤间等待

3m 46.0s

含步骤间等待

2m 59.8s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0886

OpenAI: GPT-5.5

$0.0975

OpenAI: GPT-5.5

$0.0806

OpenAI: GPT-5.5

$0.0955

OpenAI: GPT-5.5

$0.0724

OpenAI: GPT-5.5

通道指纹

来源步骤 knowledge_recall knowledge_recall knowledge_recall knowledge_recall knowledge_recall

平台 one-api (100%) unknown (0%) unknown (0%) 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）

通道	response_headers_notable	response_body_traits	signals	request_id_chain
O-Pro	server: cdn x-new-api-version: sync-upstream-main-20260630-42a06689f305 x-oneapi-request-id: 202607020712449800409548268d9d6QrROEW6k	—	hdr:x-new-api-versionhdr:x-oneapi-request-id	—
O-Pro	cf-ray: a14bd8c5287ae538-LAX server: nginx/1.27.5 set-cookie: __cf_bm=zJzbj6nyU6C9cSZ0RY1Jr6K492CQFm0FrTm1_NjZ34g-1782976395.068157-1.0.1.1-vhRj7ELnn57jQrY12glTOQIhZozkw_zo9h_dam0lRtIJ4aJFh9YeKzFQq_wAr4y0ngv0LIYxTaXUpWrsyX_t5GuZV7uDYQ0TFWcbIHmirXrTI9Y3g70Yo_uRyt	—	—	tr-ew7y4grd-kzqqmgy2tj-076311 a14bd8c5287ae538-LAX
官方基准	server: nginx/1.22.1	—	—	—
M-Pro	server: nginx	—	—	—
cx	server: GoFrame HTTP Server	—	—	—

协议指纹评分（机器自动）

总分 / 100 84 92 99 81 91

基线状态已对照基线已对照基线已对照基线已对照基线已对照基线

缓存命中比 10.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20

延迟基线 1.3 /10 ×15 6.1 /10 ×15 10.0 /10 ×15 0.7 /10 ×15 9.9 /10 ×15

模型匹配 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

系统提示纯净 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 0.0 /10 ×8

身份(结构) 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

身份(自由) 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

知识截止 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

流式投递未计入未计入未计入未计入未计入

另有 9 个维度不适用于本服务，已排除评分（不计入总分）

综合结论

与基线相似度 84 v3.26.0 92 v3.26.0 基准（参考） 81 v3.26.0 91 v3.26.0

雷达图例

共 8 个维度（顺时针，从顶部 12 点起）— 点击展开对照

1 模型匹配 model_match
2 知识截止 cutoff_match
3 自由身份 identity_free_clean
4 系统提示纯净 system_prompt_clean
5 缓存命中比 cache_hit_ratio_match
6 知识召回 knowledge_recall_match
7 延迟基线 latency_baseline_match
8 结构化身份 identity_structured_match

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 6360 ms 5605 ms 2979 ms 8144 ms 3584 ms

输出速度 51.8 tok/s - 38.8 tok/s 31.5 tok/s 45.1 tok/s

完成步骤 5/5 成功 5/5 成功 5/5 成功 5/5 成功 5/5 成功

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0886

$0.0975

$0.0806

$0.0955

$0.0724

缓存读取占比 ⓘ 76% 73% 79% 73% 62%

揭晓

通道

服务商 WanmoAPI ↗

通道O-Pro

服务商 TopRouterCN ↗

通道O-Pro

官方基准

服务商 AnyAIGC ↗

通道M-Pro

服务商 ShareAi ↗

通道cx

返回列表