模型对比 | RPDiag

测试说明点击展开 5 步 quick-probe-v1 详解

前 5 步同一会话，第 6 步独立新会话，覆盖 6 个独立信号： 1. ping/pong —— 单字指令遵循 + 建立缓存上下文。 2. 身份（结构化）—— 三行 vendor / brand / model 格式，机器可解析。 3. 知识截止 —— 跨越 5 分钟 cache 边界后追问，检测 sliding 5m cache 是否真正命中。 4. 身份（自由格式）—— 自然语言自报身份，给包装层（Kiro 等）暴露品牌的机会。 5. 世界知识层级 —— 5 道公共事件硬事实题，按答对档位映射实测 tier，与请求模型对照。 6. 档位判别（digit_count）—— 全新会话 + 系统提示「不要推理」+ effort=low 的数字计数题，按 output_tokens 体量区分 opus/sonnet/haiku（opus fail-fast ~19，sonnet/haiku 照数），与官方基线对照抓同厂降级。前 5 步之间随机延迟 1–4 分钟。第 3 步（cutoff）累计跨度 > 5 分钟保住 sliding cache 检测；总跨度 > 6 分钟。第 6 步是独立新会话（不复用前 5 轮上下文，否则 opus 不再 fail-fast）。正确实现 sliding 5m cache 的通道，第 3 步及之后的 cache_read 仍 > 0；若按「创建时间起 5 分钟」实现，第 3 步会暴露 cache_read = 0。

5 通道 · 6 步骤

2026/06/25 运行日期

claude-sonnet-4-6 官方基准

全部 ≥ 70 与基线相似度

已揭晓揭晓进度

步骤 / 指标

官方基准

Aiide

O-Max

PackyCode

cc-Zeus

duckcoding

cc专用

AiTokensFlux

O-Pro

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code Claude Code

模型 claude-sonnet-4-6 claude-sonnet-4-6 claude-sonnet-4-6 claude-sonnet-4-6 claude-sonnet-4-6

客户端 claude-cli v2.1.160 claude-cli v2.1.160 claude-cli v2.1.160 claude-cli v2.1.160 claude-cli v2.1.160

偏差标红基于官方基准

步骤 1: ping text

时间 13:55:53 ~ 13:55:58 (4.7s) 13:55:49 ~ 13:55:59 (6.9s) 13:55:47 ~ 13:55:57 (5.8s) 13:55:45 ~ 13:55:55 (5.3s) 13:55:51 ~ 13:56:06 (10.5s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,745 ms 4,318 ms 3,523 ms 3,966 ms 9,411 ms

首文本(SSE delta) 2,747 ms 4,319 ms 3,744 ms 3,977 ms 9,467 ms

总耗时 2,877 ms 4,774 ms 3,744 ms 4,267 ms 9,702 ms

输入 3 3 3 3 3

输出 5 5 5 5 5

缓存创建

32,318

5m: 0 / 1h: 32,318

32,457

5m: 0 / 1h: 32,457

32,318

5m: 0 / 1h: 32,318

32,318

5m: 0 / 1h: 32,318

1,464

5m: 0 / 1h: 1,464

缓存读取 0 0 0 0 30,993

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 13:58:43 ~ 13:58:48 (4.4s) 13:58:20 ~ 13:58:30 (5.9s) 13:57:25 ~ 13:57:35 (8.2s) 13:58:22 ~ 13:58:32 (7.3s) 13:57:49 ~ 13:57:59 (9.7s)

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) 2,908 ms 4,030 ms 5,617 ms 4,747 ms 8,429 ms

首文本(SSE delta) 2,911 ms 4,031 ms 5,644 ms 4,875 ms 8,458 ms

总耗时 3,297 ms 4,800 ms 7,038 ms 5,865 ms 8,926 ms

输入 3 3 3 3 3

输出 22 22 22 22 22

缓存创建

189

5m: 0 / 1h: 189

189

5m: 0 / 1h: 189

189

5m: 0 / 1h: 189

189

5m: 0 / 1h: 189

189

5m: 0 / 1h: 189

缓存读取 32,318 32,457 32,318 32,318 32,457

响应

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

步骤 3: cutoff text

时间 14:00:27 ~ 14:00:32 (4.5s) 13:59:58 ~ 14:00:08 (5.2s) 13:58:58 ~ 13:59:08 (5.6s) 14:01:26 ~ 14:01:31 (4.3s) 14:01:11 ~ 14:01:46 (31.4s)

提示词

查看本步 Prompt

请直接回答你的知识截止日期，格式 YYYY-MM，不要附加任何说明。

HTTP 头(TTFB) 2,932 ms 3,526 ms 4,046 ms 2,725 ms 30,430 ms

首文本(SSE delta) 3,305 ms 3,684 ms 4,050 ms 2,726 ms 30,431 ms

总耗时 3,544 ms 3,977 ms 4,508 ms 2,979 ms 30,590 ms

输入 3 3 3 3 3

输出 8 8 8 8 8

缓存创建

5m: 0 / 1h: 57

5m: 0 / 1h: 55

5m: 0 / 1h: 57

5m: 0 / 1h: 55

缓存读取 32,507 32,646 32,507 32,507 32,646

响应

2025-08

2025-08

2025-08

2025-08

2025-08

步骤 4: identity_free text

时间 14:03:01 ~ 14:03:06 (4.2s) 14:01:13 ~ 14:01:23 (6.8s) 14:02:36 ~ 14:02:41 (4.7s) 14:05:31 ~ 14:05:36 (4.3s) 14:03:48 ~ 14:03:58 (5.8s)

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) 2,037 ms 4,339 ms 2,690 ms 2,895 ms 3,881 ms

首文本(SSE delta) 2,434 ms 4,741 ms 2,691 ms 2,906 ms 3,901 ms

总耗时 3,123 ms 5,461 ms 3,673 ms 3,559 ms 4,990 ms

输入 3 3 3 3 3

输出 55 55 54 44 62

缓存创建

5m: 0 / 1h: 39

5m: 0 / 1h: 35

5m: 0 / 1h: 43

5m: 0 / 1h: 39

5m: 0 / 1h: 43

缓存读取 32,564 32,703 32,562 32,564 32,701

输出速度 79.8 tok/s 76.4 tok/s 55.0 tok/s 67.4 tok/s 56.9 tok/s

响应

我是 Claude，由 Anthropic 打造的 AI 助手。当前版本为 Claude Sonnet 4.6，专为对话、编程、分析等任务设计。

我是 Claude，由 Anthropic 开发的 AI 助手。我基于大型语言模型构建，旨在提供安全、有益的对话和任务协助。

我是 Claude，由 Anthropic 开发的 AI 助手。我能帮助回答问题、分析信息、编写代码、创作内容等多种任务。

我是 Claude，由 Anthropic 打造的 AI 助手。当前运行的模型版本是 claude-sonnet-4-6。

我是 Claude，由 Anthropic 开发的 AI 助手。我基于大型语言模型构建，能够帮助用户完成写作、编程、分析、问答等各类任务。

步骤 5: knowledge_recall text

时间 14:05:15 ~ 14:05:25 (5.0s) 14:03:57 ~ 14:04:07 (8.2s) 14:06:13 ~ 14:06:24 (5.7s) 14:07:08 ~ 14:07:18 (5.4s) 14:07:39 ~ 14:07:49 (5.4s)

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) 1,991 ms 2,981 ms 2,039 ms 1,909 ms 2,552 ms

首文本(SSE delta) 2,022 ms 3,021 ms 2,040 ms 1,913 ms 2,553 ms

总耗时 4,054 ms 5,926 ms 4,446 ms 4,442 ms 4,515 ms

输入 3 3 3 3 3

输出 205 204 205 205 204

缓存创建

8,047

5m: 0 / 1h: 8,047

8,043

5m: 0 / 1h: 8,043

8,048

5m: 0 / 1h: 8,048

8,036

5m: 0 / 1h: 8,036

2,338

5m: 0 / 1h: 2,338

缓存读取 0 0 0 0 5,718

输出速度 100.9 tok/s 70.2 tok/s 85.2 tok/s 81.1 tok/s 104.0 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

步骤 6: digit_count text

时间 14:08:57 ~ 14:09:22 (20.2s) 14:06:15 ~ 14:06:36 (15.5s) 14:09:11 ~ 14:09:26 (13.8s) 14:08:43 ~ 14:09:03 (18.4s) 14:08:53 ~ 14:09:13 (16.3s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=239839529950646606557884508478903582590511944146006821535946344507538259

HTTP 头(TTFB) 2,147 ms 2,089 ms 1,349 ms 1,177 ms 2,115 ms

首文本(SSE delta) 2,542 ms 2,091 ms 1,350 ms 1,179 ms 2,509 ms

总耗时 18,833 ms 13,958 ms 12,718 ms 17,407 ms 15,390 ms

输入 164 164 164 164 164

输出 1,399 1,058 939 1,554 1,089

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0 0

输出速度 85.9 tok/s 89.2 tok/s 82.6 tok/s 95.8 tok/s 84.5 tok/s

响应

## Counting digits in s

**s** = `239839529950646606557884508478903582590511944146006821535946344507538259`

### Length count (groups of 10):
```
2398395299 → 10
5064660655 → 10
7884508478 → 10
903582

```json
{"n":72,"c":[8,3,9]}
```

**Verification:**

`s = 239839529950646606557884508478903582590511944146006821535946344507538259`

- **Length**: 72 digits
- **0s** (×8): positions 12,17,26,32,39,49,

## Counting digits in s

**s** = `239839529950646606557884508478903582590511944146006821535946344507538259`

### Length
Grouping in blocks of 4:
`2398 3952 9950 6466 0655 7884 5084 7890 3582 5905 1194

## Counting digits in s

**s =** `239839529950646606557884508478903582590511944146006821535946344507538259`

### Length
Grouping into blocks of 10:
```
2398395299  (1–10)
5064660655  (11–20)
788450847

```json
{"n":72,"c":[8,3,9]}
```

**Verification:**

`s = 239839529950646606557884508478903582590511944146006821535946344507538259`

- **Length:** 72 digits
- **0s** (×8): positions 12,17,26,32,39,49,

总计

总输入(含缓存) 138,218 138,766 138,219 138,207 138,783

总输出 1,694 1,352 1,233 1,838 1,390

总缓存创建

40,650

5m: 0 / 1h: 40,650

40,781

5m: 0 / 1h: 40,781

40,653

5m: 0 / 1h: 40,653

40,639

5m: 0 / 1h: 40,639

4,089

5m: 0 / 1h: 4,089

总缓存读取 97,389 97,806 97,387 97,389 134,515

输出速度 87.3 tok/s 85.0 tok/s 81.2 tok/s 92.9 tok/s 85.0 tok/s

总执行时间 43.1s 48.5s 43.8s 45.0s 1m 19.1s

总墙钟时间

13m 29.0s

含步骤间等待

10m 46.8s

含步骤间等待

13m 39.0s

含步骤间等待

13m 18.0s

含步骤间等待

13m 22.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2076

Anthropic: Claude Sonnet 4.6

$0.2031

Anthropic: Claude Sonnet 4.6

$0.2007

Anthropic: Claude Sonnet 4.6

$0.2097

Anthropic: Claude Sonnet 4.6

$0.0771

Anthropic: Claude Sonnet 4.6

通道指纹

来源步骤 digit_count digit_count digit_count digit_count digit_count

平台 direct (100%) one-api (100%) one-api (100%) one-api (100%) direct (100%)

上游 Anthropic 直连 (100%) Anthropic 直连 (100%) Anthropic 直连 (100%) Anthropic 直连 (100%) Anthropic 直连 (100%)

CDN - cloudflare cloudflare - -

ID 格式 msg_01HbowBawkPge2iv8okJhEnY msg_014Bem83dNZuF5nELVbuqwYu msg_01185wP2BW3Y9HcxBcd8uySC msg_0137BooCQZ5NVWkthfdk6jcH msg_01QLKL9zn1MJXs421rXh4qA8

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
官方基准	server: nginx/1.22.1	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: false usage_has_cache_fields: true	body:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	req_011CcPUkMRxpw9y6afzpT5gv
O-Max	cf-ray: a111c9100ce35811-AMS server: cloudflare x-new-api-version: 0.12.11 x-oneapi-request-id: 202606250606178201162508268d9d6ktPJFM2n	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: false usage_has_cache_fields: true	hdr:x-new-api-versionhdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a111c9100ce35811-AMS
cc-Zeus	via: 1.1 Caddy cf-ray: a111cd544b70cb93-LAX server: cloudflare x-oneapi-request-id: 20260625140912385946006iAKmOubR	id_format: msg_prefix via_chain: ["1.1 Caddy"] extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: false usage_has_cache_fields: true	hdr:x-oneapi-request-idhdr:server:cloudflarehdr:cf-rayhdr:viabody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a111cd544b70cb93-LAX
cc专用	server: Nginx x-oneapi-request-id: 202606250608443229432018268d9d6EmswoLOF	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: false usage_has_cache_fields: true	hdr:x-oneapi-request-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
O-Pro	server: nginx	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: true has_inference_geo: true model_has_date_suffix: false usage_has_cache_fields: true	body:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—

协议指纹评分（机器自动）

总分 / 100 100 97 98 98 94

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

缓存 TTL 一致 10.0 /10 ×15 10.0 /10 ×15 10.0 /10 ×15 10.0 /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 10.0 /10 ×14

缓存连续性 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

sliding 5m 10.0 /10 ×13 10.0 /10 ×13 10.0 /10 ×13 10.0 /10 ×13 10.0 /10 ×13

原生 msg-ID 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8

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

知识截止 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

service_tier 10.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6

inference_geo 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

延迟基线 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 0.0 /10 ×5

流式投递 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

Req-ID 透传 10.0 /10 ×4 0.0 /10 ×4 0.0 /10 ×4 0.0 /10 ×4 0.0 /10 ×4

stop_reason 10.0 /10 ×3 10.0 /10 ×3 10.0 /10 ×3 10.0 /10 ×3 10.0 /10 ×3

SDK 一致 10.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2 10.0 /10 ×2

综合结论

与基线相似度基准（参考） 97 v3.24.1 98 v3.24.1 98 v3.24.1 94 v3.24.1

雷达图例

共 21 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 SDK 一致 sdk_consistency
4 流式投递 buffer_dump_match
5 自由身份 identity_free_clean
6 停止原因 stop_reason_present
7 系统提示纯净 system_prompt_clean
8 Service Tier service_tier_present
9 缓存命中比 cache_hit_ratio_match
10 缓存 TTL 一致 cache_ttl_consistency
11 推理地区 inference_geo_present
12 缓存连续 cache_continuity_intra
13 知识召回 knowledge_recall_match
14 延迟基线 latency_baseline_match
15 原生 msg-ID anthropic_msg_id_format
16 Sliding 缓存 cache_sliding_correctness
17 结构化身份 identity_structured_match
18 信封自报一致 envelope_self_report_match
19 中文遵循 instruction_following_lang
20 档位思考量 tier_thinking_volume_match
21 Req-ID 透传 anthropic_request_id_passthrough

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 5955 ms 6483 ms 6021 ms 6420 ms 12352 ms

输出速度 87.3 tok/s 85.0 tok/s 81.2 tok/s 92.9 tok/s 85.0 tok/s

完成步骤 6/6 成功 6/6 成功 6/6 成功 6/6 成功 6/6 成功

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2076

$0.2031

$0.2007

$0.2097

$0.0771

缓存读取占比 ⓘ 70% 70% 70% 70% 97%

揭晓

通道

官方基准

服务商 Aiide ↗

通道O-Max

服务商 PackyCode ↗

通道cc-Zeus

服务商 duckcoding ↗

通道cc专用

服务商 AiTokensFlux ↗

通道O-Pro

返回列表