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。

4 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 另含 1 任务失败

已揭晓揭晓进度

⚠ 1 通道不可测 · 已从主对比表移除点击展开原因

100X / O-Puls/Pro
```
评测失败
```

步骤 / 指标

lucen.cc

O-Plus/Pro

OmniaKey

O-Pro

right

o-cx

官方基准

基本信息

服务名 Codex Codex Codex Codex

模型 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

偏差标红基于官方基准

步骤 1: ping text

时间 12:11:25 ~ 12:11:45 (17.6s) 12:11:23 ~ 12:11:43 (15.6s) 12:11:21 ~ 12:11:36 (12.9s) 12:11:29 ~ 12:11:44 (12.1s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 6,232 ms 4,138 ms 627 ms 844 ms

总耗时 6,644 ms 4,743 ms 2,377 ms 1,400 ms

输入 8,201 1,548 741 1,548

输出 5 5 782 5

缓存读取 0 8,576 7,040 8,576

输出速度 - - burst -

响应

pong

pong

pong

pong

步骤 2: identity text

时间 12:12:22 ~ 12:12:42 (19.0s) 12:12:15 ~ 12:12:35 (15.6s) 12:12:09 ~ 12:12:24 (13.4s) 12:12:13 ~ 12:12:28 (13.9s)

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。自报身份，输出严格遵守以下 3 行格式（不允许 markdown、代码块、额外文本）：
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) 4,802 ms 2,480 ms 921 ms 873 ms

总耗时 8,127 ms 4,832 ms 2,916 ms 3,214 ms

输入 334 153 1,275 153

输出 266 119 23 98

缓存读取 8,008 10,112 6,656 10,112

输出速度 80.0 tok/s 50.6 tok/s - 41.9 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-codex

vendor: openai
brand: codex
model: gpt-5

步骤 3: cutoff text

时间 12:13:00 ~ 12:13:20 (16.4s) 12:12:53 ~ 12:13:08 (13.8s) 12:13:03 ~ 12:13:23 (15.7s) 12:12:55 ~ 12:13:10 (12.8s)

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) 4,638 ms 2,682 ms 1,743 ms 997 ms

总耗时 5,276 ms 2,974 ms 5,126 ms 2,086 ms

输入 336 1,725 1,946 199

输出 8 8 8 8

缓存读取 8,052 8,576 8,576 10,112

响应

2024-06

2024-06

2024-06

2024-06

步骤 4: identity_free text

时间 12:13:48 ~ 12:14:08 (15.8s) 12:13:41 ~ 12:13:56 (13.7s) 12:13:54 ~ 12:14:09 (12.9s) 12:13:31 ~ 12:13:46 (14.2s)

提示词

查看本步 Prompt

请用中文介绍你自己是什么产品或助手，由谁开发，50 字以内。

HTTP 头(TTFB) 4,392 ms 2,151 ms - 950 ms

总耗时 4,997 ms 3,045 ms - 3,570 ms

输入 337 221 - 234

输出 29 34 - 33

缓存读取 8,083 10,112 - 10,112

输出速度 - 38.0 tok/s - 12.6 tok/s

响应

我是 OpenAI 开发的通用 AI 助手，可帮助回答问题、写作、编程与分析。

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

我是 OpenAI 开发的 Codex 编程助手，可协助阅读、修改、测试代码，并解答技术问题。

我是 OpenAI 开发的 Codex，一款基于 GPT-5 的编程助手，可协助阅读、修改、测试代码。

步骤 5: knowledge_recall text

时间 12:14:38 ~ 12:15:08 (25.5s) 12:14:17 ~ 12:14:37 (18.5s) 12:14:40 ~ 12:15:10 (26.4s) 12:14:27 ~ 12:14:47 (15.6s)

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) 8,565 ms 2,549 ms 952 ms 839 ms

总耗时 14,505 ms 6,376 ms 15,570 ms 4,906 ms

输入 8,971 8,126 8,422 8,126

输出 681 182 681 185

缓存读取 0 0 0 0

输出速度 114.6 tok/s 47.6 tok/s 46.6 tok/s 45.5 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":"UNKNOWN","euro2024":"UNKNOWN","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":"UNKNOWN","euro2024":"UNKNOWN","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) 42,322 49,149 34,656 49,172

总输出 989 348 1,494 329

总缓存创建 0 0 0 0

总缓存读取 24,143 37,376 22,272 38,912

输出速度 102.2 tok/s 47.4 tok/s 46.6 tok/s 35.0 tok/s

总执行时间 1m 34.2s 1m 17.2s 1m 21.3s 1m 8.7s

总墙钟时间

3m 42.9s

含步骤间等待

3m 13.9s

含步骤间等待

3m 48.9s

含步骤间等待

3m 17.9s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1326

OpenAI: GPT-5.5

$0.0880

OpenAI: GPT-5.5

$0.1179

OpenAI: GPT-5.5

$0.0806

OpenAI: GPT-5.5

通道指纹

来源步骤 knowledge_recall knowledge_recall knowledge_recall knowledge_recall

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

上游 - - - -

CDN cloudflare cloudflare 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-Plus/Pro	cf-ray: a14ad32b287266ec-AMS server: cloudflare	—	hdr:server:cloudflarehdr:cf-ray	bc78a575-bc31-49d7-a478-3b1391af5241 a14ad32b287266ec-AMS
O-Pro	cf-ray: a14ad2ae9fdf7a96-AMS server: cloudflare x-oneapi-request-id: 202607020414208624077668268d9d6GVxbTbQq	—	hdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-ray	a14ad2ae9fdf7a96-AMS
o-cx	via: 1.1 Caddy cf-ray: a14ad335a8cee393-NRT server: cloudflare	via_chain: ["1.1 Caddy"]	hdr:server:cloudflarehdr:cf-rayhdr:via	a14ad335a8cee393-NRT
官方基准	server: nginx/1.22.1	—	—	—

协议指纹评分（机器自动）

总分 / 100 68 93 81 100

基线状态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

延迟基线 1.6 /10 ×15 5.8 /10 ×15 9.3 /10 ×15 10.0 /10 ×15

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

系统提示纯净 0.0 /10 ×8 10.0 /10 ×8 0.0 /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

身份(自由) 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

流式投递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 68 v3.26.0 93 v3.26.0 81 v3.26.0 基准（参考）

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 7910 ms 4394 ms 5198 ms 3035 ms

输出速度 102.2 tok/s 47.4 tok/s 46.6 tok/s 35.0 tok/s

完成步骤 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 ⓘ

$0.1326

$0.0880

$0.1179

$0.0806

缓存读取占比 ⓘ 57% 76% 64% 79%

揭晓

通道

服务商 lucen.cc ↗

通道O-Plus/Pro

服务商 OmniaKey ↗

通道O-Pro

服务商 right ↗

通道o-cx

官方基准

返回列表