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

全部 ≥ 70 与基线相似度

已揭晓揭晓进度

步骤 / 指标

JuCodex

O-Pro

SSSAiCode

O-Team/Plus

官方基准

5Cookie

O-Pro

SSSAiCode

O-Team/Plus-L2

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 17:00:37 ~ 17:00:57 (16.1s) 17:00:41 ~ 17:01:06 (24.0s) 17:00:45 ~ 17:01:00 (12.7s) 17:00:39 ~ 17:00:54 (13.5s) 17:00:43 ~ 17:01:08 (20.3s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 4,542 ms 7,143 ms 929 ms 981 ms 5,112 ms

总耗时 5,272 ms 10,505 ms 2,059 ms 2,470 ms 6,885 ms

输入 1,548 1,548 1,548 1,548 1,548

输出 5 5 5 5 5

缓存读取 8,576 8,576 8,576 8,576 8,576

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 17:01:20 ~ 17:01:40 (18.0s) 17:01:49 ~ 17:02:09 (18.6s) 17:01:17 ~ 17:01:37 (17.9s) 17:01:10 ~ 17:01:30 (18.1s) 17:01:32 ~ 17:01:52 (16.3s)

提示词

查看本步 Prompt

以下是对你自身的 metadata 探针。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。按以下 3 行回答自己的身份，每行一个 label，不要寒暄不要解释：
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) 2,399 ms 1,385 ms 3,646 ms 2,723 ms 1,203 ms

总耗时 6,969 ms 5,123 ms 7,211 ms 7,214 ms 5,856 ms

输入 1,693 1,693 157 157 1,693

输出 102 155 132 160 20

缓存读取 8,576 8,576 10,112 10,112 8,576

输出速度 22.3 tok/s 41.5 tok/s 37.0 tok/s 35.6 tok/s -

响应

vendor: openai
brand: codex
model: gpt-5

vendor: openai
brand: codex
model: gpt-5

vendor: openai
brand: codex
model: gpt-5

vendor: openai
brand: codex
model: gpt-5

vendor: openai
brand: codex
model: gpt-5

步骤 3: cutoff text

时间 17:02:08 ~ 17:02:23 (14.2s) 17:02:50 ~ 17:03:10 (16.0s) 17:01:58 ~ 17:02:13 (13.1s) 17:02:03 ~ 17:02:18 (12.5s) 17:02:36 ~ 17:02:56 (16.7s)

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) 2,540 ms 1,394 ms 1,402 ms 857 ms 1,170 ms

总耗时 3,346 ms 5,595 ms 2,436 ms 1,774 ms 3,244 ms

输入 203 201 8,393 193 193

输出 8 8 8 8 8

缓存读取 10,112 10,112 1,920 10,112 10,112

响应

2024-06

2024-06

2024-06

2024-06

2024-06

步骤 4: identity_free text

时间 17:02:47 ~ 17:03:07 (16.5s) 17:03:52 ~ 17:04:07 (13.6s) 17:02:45 ~ 17:03:00 (12.9s) 17:02:34 ~ 17:02:49 (13.5s) 17:03:29 ~ 17:03:44 (13.6s)

提示词

查看本步 Prompt

用中文说说你的身份：什么产品？由谁打造？50 字以内。

HTTP 头(TTFB) 4,408 ms 1,979 ms 860 ms - 1,777 ms

总耗时 5,841 ms 3,211 ms 2,211 ms - 3,213 ms

输入 235 1,769 1,768 - 1,758

输出 34 31 24 - 34

缓存读取 10,112 8,576 8,576 - 8,576

输出速度 23.7 tok/s 25.2 tok/s - - 23.7 tok/s

响应

我是 OpenAI 打造的 Codex，基于 GPT-5 的编码助手，用于协助编程、调试和工程任务。

我是 OpenAI 开发的 Codex 编程助手，专注于代码理解、修改、调试与协作开发。

我叫 Codex，属于 OpenAI 的编程智能体产品，由 OpenAI 开发。

我是 OpenAI 开发的 Codex 编程助手，可协助阅读代码、修改项目、运行命令并排查问题。

我是 OpenAI 开发的 Codex，基于 GPT-5 的编程助手，用于协助代码、调试和工程任务。

步骤 5: knowledge_recall text

时间 17:03:44 ~ 17:04:29 (44.5s) 17:04:27 ~ 17:04:47 (15.3s) 17:03:17 ~ 17:03:37 (15.9s) 17:03:19 ~ 17:03:39 (15.4s) 17:04:14 ~ 17:04:34 (19.6s)

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) 19,632 ms 1,187 ms 1,127 ms 831 ms 2,626 ms

总耗时 33,772 ms 4,889 ms 5,196 ms 4,734 ms 9,191 ms

输入 8,126 8,126 8,126 8,126 8,126

输出 679 174 179 179 179

缓存读取 0 0 0 0 0

输出速度 48.0 tok/s 47.0 tok/s 44.0 tok/s 45.9 tok/s 27.3 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) 49,181 49,177 49,176 38,824 49,158

总输出 828 373 348 352 246

总缓存创建 0 0 0 0 0

总缓存读取 37,376 35,840 29,184 28,800 35,840

输出速度 40.5 tok/s 41.5 tok/s 40.7 tok/s 40.4 tok/s 26.6 tok/s

总执行时间 1m 49.3s 1m 27.6s 1m 12.5s 1m 13.0s 1m 26.5s

总墙钟时间

3m 52.0s

含步骤间等待

4m 5.9s

含步骤间等待

2m 51.9s

含步骤间等待

2m 59.9s

含步骤间等待

3m 51.1s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1026

OpenAI: GPT-5.5

$0.0958

OpenAI: GPT-5.5

$0.1250

OpenAI: GPT-5.5

$0.0751

OpenAI: GPT-5.5

$0.0919

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

上游 - - - - -

CDN cloudflare - - - -

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	cf-ray: a14c7aa5a9194ba8-FRA server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 20260702090347155877320m0NEbUZJ	—	hdr:x-new-api-versionhdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-ray	a14c7aa5a9194ba8-FRA
O-Team/Plus	via: 1.1 Caddy server: Photon-Edge	via_chain: ["1.1 Caddy"]	hdr:via	1782983068311-pl0qkqy8q0o1b
官方基准	server: nginx/1.22.1	—	—	—
O-Pro	via: 1.1 Caddy x-oneapi-request-id: 202607020903204126857228268d9d6MH3qhv9S	via_chain: ["1.1 Caddy"]	hdr:x-oneapi-request-idhdr:via	—
O-Team/Plus-L2	via: 1.1 Caddy server: Photon-Edge	via_chain: ["1.1 Caddy"]	hdr:via	1782983055426-7if9w73z3tipv

协议指纹评分（机器自动）

总分 / 100 92 93 100 100 98

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 5.2 /10 ×15 7.3 /10 ×15 10.0 /10 ×15 10.0 /10 ×15 9.1 /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 10.0 /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 92 v3.26.0 93 v3.26.0 基准（参考） 100 v3.26.0 98 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 11040 ms 5865 ms 3823 ms 3238 ms 5678 ms

输出速度 40.5 tok/s 41.5 tok/s 40.7 tok/s 40.4 tok/s 26.6 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 ⓘ

$0.1026

$0.0958

$0.1250

$0.0751

$0.0919

缓存读取占比 ⓘ 76% 73% 59% 74% 73%

揭晓

通道

服务商 JuCodex ↗

通道O-Pro

服务商 SSSAiCode ↗

通道O-Team/Plus

官方基准

服务商 5Cookie ↗

通道O-Pro

服务商 SSSAiCode ↗

通道O-Team/Plus-L2

返回列表