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-5.4 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-L2

SSSAiCode

O-Team/Plus

官方基准

OmniaKey

O-Pro

基本信息

服务名 Codex Codex Codex Codex Codex

模型 gpt-5.4 gpt-5.4 gpt-5.4 gpt-5.4 gpt-5.4

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 20:47:02 ~ 20:47:17 (14.8s) 20:47:08 ~ 20:47:23 (12.9s) 20:47:06 ~ 20:47:21 (13.7s) 20:47:10 ~ 20:47:25 (13.3s) 20:47:04 ~ 20:47:19 (14.0s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,177 ms 1,555 ms 2,043 ms 940 ms 2,400 ms

总耗时 3,859 ms 2,450 ms 3,196 ms 2,609 ms 2,913 ms

输入 1,699 163 1,699 6,819 163

输出 5 5 5 5 5

缓存读取 7,040 8,576 7,040 1,920 8,576

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 20:47:35 ~ 20:48:00 (21.8s) 20:48:08 ~ 20:48:28 (15.5s) 20:48:04 ~ 20:48:24 (19.6s) 20:47:40 ~ 20:48:00 (17.4s) 20:48:06 ~ 20:48:31 (22.2s)

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) 2,036 ms 1,229 ms 1,539 ms 891 ms 3,278 ms

总耗时 10,874 ms 5,017 ms 9,084 ms 6,689 ms 10,841 ms

输入 6,952 296 296 6,952 296

输出 157 161 369 275 383

缓存读取 1,920 8,576 8,576 1,920 8,576

输出速度 17.8 tok/s 42.5 tok/s 48.9 tok/s 47.4 tok/s 50.6 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 20:48:27 ~ 20:48:47 (16.1s) 20:49:04 ~ 20:49:20 (12.8s) 20:48:57 ~ 20:49:12 (13.4s) 20:48:34 ~ 20:48:49 (12.1s) 20:48:53 ~ 20:49:08 (14.3s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 2,683 ms 1,267 ms 1,510 ms 811 ms 2,687 ms

总耗时 4,343 ms 2,370 ms 2,998 ms 1,470 ms 3,451 ms

输入 332 340 342 1,878 332

输出 36 27 8 8 27

缓存读取 8,576 8,576 8,576 7,040 8,576

输出速度 21.7 tok/s - - - -

响应

2024-06

2024-06

2024-06

2024-06

2024-06

步骤 4: identity_free text

时间 20:49:11 ~ 20:49:31 (15.2s) 20:49:58 ~ 20:50:13 (12.8s) 20:49:50 ~ 20:50:06 (13.3s) 20:49:06 ~ 20:49:22 (13.5s) 20:49:27 ~ 20:49:48 (15.9s)

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) 2,819 ms 1,228 ms 1,516 ms 1,358 ms 2,614 ms

总耗时 4,251 ms 2,359 ms 2,816 ms 2,845 ms 3,888 ms

输入 364 372 1,907 371 361

输出 29 28 13 28 13

缓存读取 8,576 8,576 7,040 8,576 8,576

响应

我是 OpenAI 打造的 Codex 助手。

我是 OpenAI 打造的 Codex 助手。

我是 Codex，由 OpenAI 开发。

我是 ChatGPT，由 OpenAI 开发。

我是 Codex，由 OpenAI 开发。

步骤 5: knowledge_recall text

时间 20:50:00 ~ 20:50:25 (21.5s) 20:50:43 ~ 20:51:03 (16.3s) 20:50:48 ~ 20:51:08 (16.4s) 20:49:53 ~ 20:50:13 (15.5s) 20:50:25 ~ 20:50:45 (17.9s)

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) 2,189 ms 1,682 ms 2,031 ms 854 ms 3,335 ms

总耗时 9,761 ms 5,877 ms 5,882 ms 4,807 ms 7,025 ms

输入 6,741 6,741 6,741 6,741 85

输出 380 179 179 179 179

缓存读取 0 0 0 0 6,656

输出速度 50.2 tok/s 42.7 tok/s 46.5 tok/s 45.3 tok/s 48.5 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":"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) 42,200 42,216 42,217 42,217 42,197

总输出 607 400 574 495 607

总缓存创建 0 0 0 0 0

总缓存读取 26,112 34,304 31,232 19,456 40,960

输出速度 31.7 tok/s 42.6 tok/s 48.1 tok/s 46.6 tok/s 49.9 tok/s

总执行时间 1m 29.5s 1m 10.3s 1m 16.3s 1m 11.8s 1m 24.4s

总墙钟时间

3m 22.9s

含步骤间等待

3m 55.1s

含步骤间等待

4m 1.9s

含步骤间等待

3m 2.7s

含步骤间等待

3m 41.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0559

OpenAI: GPT-5.4

$0.0344

OpenAI: GPT-5.4

$0.0439

OpenAI: GPT-5.4

$0.0692

OpenAI: GPT-5.4

$0.0224

OpenAI: GPT-5.4

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%) one-api (100%)

上游 - - - - -

CDN 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-Pro	cf-ray: a14dc61a78729f35-FRA server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 20260702125002323993108mMhzXOC1	—	hdr:x-new-api-versionhdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-ray	a14dc61a78729f35-FRA
O-Team/Plus-L2	via: 1.1 Caddy server: Photon-Edge	via_chain: ["1.1 Caddy"]	hdr:via	1782996643966-h4v9g8g727lti
O-Team/Plus	via: 1.1 Caddy server: Photon-Edge	via_chain: ["1.1 Caddy"]	hdr:via	1782996649121-hmzw7akujosmj
官方基准	server: nginx/1.22.1	—	—	—
O-Pro	cf-ray: a14dc6b258b7bd8d-AMS server: cloudflare x-oneapi-request-id: 202607021250269815563458268d9d6TL9jGFNn	—	hdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-ray	a14dc6b258b7bd8d-AMS

协议指纹评分（机器自动）

总分 / 100 88 100 98 89 92

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 5.2 /10 ×20 10.0 /10 ×20

延迟基线 7.1 /10 ×15 9.9 /10 ×15 9.0 /10 ×15 10.0 /10 ×15 5.5 /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 88 v3.26.0 100 v3.26.0 98 v3.26.0 基准（参考） 92 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 6618 ms 3615 ms 4795 ms 3684 ms 5624 ms

输出速度 31.7 tok/s 42.6 tok/s 48.1 tok/s 46.6 tok/s 49.9 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 ⓘ

$0.0559

$0.0344

$0.0439

$0.0692

$0.0224

缓存读取占比 ⓘ 62% 81% 74% 46% 97%

揭晓

通道

服务商 JuCodex ↗

通道O-Pro

服务商 SSSAiCode ↗

通道O-Team/Plus-L2

服务商 SSSAiCode ↗

通道O-Team/Plus

官方基准

服务商 OmniaKey ↗

通道O-Pro

返回列表