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

1 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

NowCoding / O-Plus/Team
```
评测失败
```

步骤 / 指标

LinkAPI

O-Team

oukapi

M-Plus

官方基准

0-0

O-Team/Plus

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 14:26:58 ~ 14:27:13 (12.5s) 14:27:00 ~ 14:27:25 (22.5s) 14:27:04 ~ 14:27:19 (13.0s) 14:26:56 ~ 14:27:06 (9.4s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 1,559 ms 6,464 ms 1,047 ms 1,528 ms

总耗时 1,844 ms 12,000 ms 2,342 ms 5,784 ms

输入 1,548 1,548 1,548 8,204

输出 5 5 5 5

缓存读取 8,576 8,576 8,576 1,920

响应

pong

pong

pong

pong

步骤 2: identity text

时间 14:27:51 ~ 14:28:01 (5.5s) 14:27:53 ~ 14:28:18 (22.7s) 14:28:04 ~ 14:28:19 (13.3s) 14:27:49 ~ 14:28:09 (15.1s)

提示词

查看本步 Prompt

输出 3 行 metadata 描述你自己（当前 API 模型）。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。禁止任何额外字符：
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) 3,539 ms 8,223 ms 951 ms 1,616 ms

总耗时 4,786 ms 12,222 ms 2,473 ms 4,058 ms

输入 148 1,684 148 1,684

输出 21 20 23 20

缓存读取 10,112 8,576 10,112 8,576

响应

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

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 14:28:44 ~ 14:28:59 (13.3s) 14:28:51 ~ 14:29:11 (18.8s) 14:28:49 ~ 14:29:04 (12.5s) 14:28:47 ~ 14:29:12 (20.8s)

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) 2,092 ms 4,661 ms 920 ms 1,938 ms

总耗时 2,465 ms 8,324 ms 1,839 ms 7,377 ms

输入 1,731 194 8,387 192

输出 8 8 8 8

缓存读取 8,576 10,112 1,920 10,112

响应

2024-06

2024-06

2024-06

2024-06

步骤 4: identity_free text

时间 14:29:40 ~ 14:29:55 (13.2s) 14:29:47 ~ 14:30:12 (20.5s) 14:29:21 ~ 14:29:36 (13.0s) 14:29:32 ~ 14:29:52 (16.9s)

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) 1,538 ms 5,014 ms 1,012 ms 1,617 ms

总耗时 2,551 ms 10,084 ms 2,372 ms 5,362 ms

输入 1,763 229 227 227

输出 25 33 30 35

缓存读取 8,576 10,112 10,112 10,112

输出速度 - 6.5 tok/s 22.1 tok/s 9.3 tok/s

响应

我是 OpenAI 打造的 Codex，基于 GPT-5 的编程协作助手。

我是 OpenAI 开发的 Codex，基于 GPT-5 的编程助手，可协助代码理解、修改与调试。

我是 OpenAI 开发的 Codex 编程助手，可协助阅读、修改、调试代码并解释技术问题。

我是 Codex，基于 GPT-5 的编程助手，由 OpenAI 开发，专注代码理解、修改与调试协作。

步骤 5: knowledge_recall text

时间 14:30:33 ~ 14:30:53 (18.8s) 14:30:53 ~ 14:31:18 (23.9s) 14:29:51 ~ 14:30:11 (15.4s) 14:30:35 ~ 14:31:05 (27.2s)

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) 1,438 ms 9,237 ms 874 ms 1,643 ms

总耗时 8,080 ms 13,424 ms 4,733 ms 15,785 ms

输入 8,126 8,126 8,126 8,126

输出 179 171 182 179

缓存读取 0 0 0 0

输出速度 26.9 tok/s 40.8 tok/s 47.2 tok/s 12.7 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

总计

总输入(含缓存) 49,156 49,157 49,156 49,153

总输出 238 237 248 247

总缓存创建 0 0 0 0

总缓存读取 35,840 37,376 30,720 30,720

输出速度 26.9 tok/s 22.0 tok/s 40.6 tok/s 12.0 tok/s

总执行时间 1m 3.3s 1m 48.4s 1m 7.3s 1m 29.4s

总墙钟时间

3m 55.2s

含步骤间等待

4m 18.9s

含步骤间等待

3m 7.9s

含步骤间等待

4m 9.2s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0916

OpenAI: GPT-5.5

$0.0847

OpenAI: GPT-5.5

$0.1150

OpenAI: GPT-5.5

$0.1149

OpenAI: GPT-5.5

通道指纹

来源步骤 knowledge_recall knowledge_recall knowledge_recall knowledge_recall

平台 one-api (100%) unknown (0%) unknown (0%) 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-Team	server: nginx set-cookie: server_name_session=e1a4d69078c70b9351103fe56cdd3fd2; Max-Age=86400; httponly; path=/ x-new-api-version: v1.0.0-rc.15 x-oneapi-request-id: 202607020630342145737618268d9d6Lze87LVr	—	hdr:x-new-api-versionhdr:x-oneapi-request-id	—
M-Plus	server: nginx/1.28.1	—	—	f5fdf3bd-7627-4106-a91b-0c8f72f15441
官方基准	server: nginx/1.22.1	—	—	—
O-Team/Plus	via: 1.1 Caddy cf-ray: a14b9a5429965e53-IAD server: uvicorn, cloudflare set-cookie: __cf_bm=6j1FtFDiCgZmmdhiD9lXNmbasbr7bM..S_7295RWISo-1782973837.4632404-1.0.1.1-b3zLOlXH4dP8YAnHan7Xu5Kl3wzPedoFR0iEU3xL.ZRw_3Yo8Dw.QPgbDk6biI1.9LvBX0OQEWNdJ_86YrUlS3J7WHhQjiMz8aPxnvldjjOWcgpkUlInoPNsI	via_chain: ["1.1 Caddy"]	hdr:server:cloudflarehdr:cf-rayhdr:via	1d0a45bf1454bb343d54a3faf23ca784 a14b9a5429965e53-IAD

协议指纹评分（机器自动）

总分 / 100 98 80 99 93

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 9.7 /10 ×20 9.7 /10 ×20

延迟基线 8.7 /10 ×15 0.0 /10 ×15 10.0 /10 ×15 6.6 /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 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

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 98 v3.26.0 80 v3.26.0 基准（参考） 93 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 3945 ms 11211 ms 2752 ms 7673 ms

输出速度 26.9 tok/s 22.0 tok/s 40.6 tok/s 12.0 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 ⓘ

$0.0916

$0.0847

$0.1150

$0.1149

缓存读取占比 ⓘ 73% 76% 62% 62%

揭晓

通道

服务商 LinkAPI ↗

通道O-Team

服务商 oukapi ↗

通道M-Plus

官方基准

服务商 0-0 ↗

通道O-Team/Plus

返回列表