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

2026/06/30 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

ClaudeZZ

R-kiro

官方基准

Aiide

O-Max

AiTokensFlux

O-Pro

LingxiCode

M-

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 02:06:12 ~ 02:06:22 (6.0s) 02:06:17 ~ 02:06:23 (3.7s) 02:06:15 ~ 02:06:26 (8.1s) 02:06:13 ~ 02:06:19 (4.1s) 02:06:10 ~ 02:06:20 (7.4s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 4,478 ms 2,465 ms 6,797 ms 2,609 ms 3,381 ms

首文本(SSE delta) 4,480 ms 2,467 ms 6,798 ms 2,611 ms 3,382 ms

总耗时 4,842 ms 2,568 ms 6,893 ms 2,826 ms 3,790 ms

输入 3 3 3 3 3

输出 5 5 5 5 5

缓存创建

21,985

5m: 0 / 1h: 21,985

32,318

5m: 0 / 1h: 32,318

32,457

5m: 0 / 1h: 32,457

32,323

5m: 0 / 1h: 32,323

32,319

5m: 0 / 1h: 32,319

缓存读取 0 0 0 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 02:07:56 ~ 02:08:01 (4.8s) 02:08:25 ~ 02:08:30 (3.4s) 02:07:45 ~ 02:07:55 (6.0s) 02:08:20 ~ 02:08:30 (8.3s) 02:09:21 ~ 02:09:26 (4.1s)

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) 3,162 ms 1,907 ms 3,675 ms 6,476 ms 2,760 ms

首文本(SSE delta) 3,163 ms 1,908 ms 3,713 ms 6,477 ms 2,760 ms

总耗时 3,891 ms 2,295 ms 4,814 ms 6,806 ms 3,344 ms

输入 3 3 3 3 3

输出 22 22 22 22 22

缓存创建

209

5m: 0 / 1h: 209

202

5m: 0 / 1h: 202

202

5m: 0 / 1h: 202

202

5m: 0 / 1h: 202

202

5m: 0 / 1h: 202

缓存读取 21,985 32,318 32,457 32,323 32,319

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 02:10:02 ~ 02:10:12 (5.9s) 02:11:01 ~ 02:11:06 (3.5s) 02:10:00 ~ 02:10:10 (7.9s) 02:12:25 ~ 02:12:30 (4.2s) 02:10:35 ~ 02:10:40 (3.9s)

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) 4,321 ms 2,175 ms 6,515 ms 3,079 ms 2,613 ms

首文本(SSE delta) 4,356 ms 2,302 ms 6,648 ms 3,125 ms 2,614 ms

总耗时 4,965 ms 2,399 ms 6,679 ms 3,424 ms 3,083 ms

输入 3 3 3 3 3

输出 8 8 8 8 8

缓存创建

5m: 0 / 1h: 79

5m: 0 / 1h: 55

5m: 0 / 1h: 57

5m: 0 / 1h: 38

缓存读取 22,194 32,520 32,659 32,525 32,521

响应

2025-08

2025-08

2025-08

2025-08

2025-08

步骤 4: identity_free text

时间 02:13:43 ~ 02:13:53 (5.6s) 02:14:02 ~ 02:14:07 (3.7s) 02:12:32 ~ 02:12:42 (5.7s) 02:14:55 ~ 02:15:00 (4.4s) 02:13:35 ~ 02:13:45 (5.1s)

提示词

查看本步 Prompt

中文回答：你叫什么、属于哪个产品、谁开发？50 字内。

HTTP 头(TTFB) 4,022 ms 1,864 ms 3,776 ms 2,677 ms 3,174 ms

首文本(SSE delta) 4,025 ms 1,866 ms 3,777 ms 2,678 ms 3,175 ms

总耗时 4,781 ms 2,611 ms 4,501 ms 3,578 ms 4,068 ms

输入 3 3 3 3 3

输出 53 46 48 57 41

缓存创建

5m: 0 / 1h: 51

5m: 0 / 1h: 39

5m: 0 / 1h: 41

5m: 0 / 1h: 43

5m: 0 / 1h: 41

缓存读取 22,273 32,575 32,716 32,582 32,559

输出速度 70.1 tok/s 61.7 tok/s 66.3 tok/s 63.3 tok/s 45.9 tok/s

响应

我叫 Claude，是 Anthropic 开发的 AI 助手，目前运行的模型是 claude-sonnet-4-6，属于 Claude 产品系列。

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

我叫 Claude，属于 Anthropic 开发的 Claude 系列 AI 助手产品，当前版本为 claude-sonnet-4-6。

我是 Claude，由 Anthropic 开发的 AI 助手。当前版本为 claude-sonnet-4-6，擅长对话、编程、分析和写作等任务。

我叫 Claude，是由 Anthropic 开发的 AI 助手，当前版本为 Claude Sonnet 4.6。

步骤 5: knowledge_recall text

时间 02:17:38 ~ 02:17:48 (7.0s) 02:15:37 ~ 02:15:42 (4.7s) 02:15:33 ~ 02:15:43 (5.4s) 02:16:07 ~ 02:16:12 (4.9s) 02:17:03 ~ 02:17:13 (6.1s)

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) 3,867 ms 1,788 ms 2,467 ms 1,632 ms 2,579 ms

首文本(SSE delta) 3,868 ms 1,791 ms 2,468 ms 1,633 ms 2,581 ms

总耗时 6,134 ms 3,654 ms 4,214 ms 4,030 ms 5,282 ms

输入 3 3 3 3 3

输出 204 205 205 204 204

缓存创建

6,123

5m: 0 / 1h: 6,123

8,049

5m: 0 / 1h: 8,049

8,055

5m: 0 / 1h: 8,055

8,066

5m: 0 / 1h: 8,066

8,030

5m: 0 / 1h: 8,030

缓存读取 2,136 0 0 0 0

输出速度 90.0 tok/s 110.0 tok/s 117.4 tok/s 85.1 tok/s 75.5 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

步骤 6: digit_count text

时间 02:20:44 ~ 02:21:04 (17.1s) 02:17:06 ~ 02:17:21 (11.4s) 02:19:37 ~ 02:19:57 (18.7s) 02:19:26 ~ 02:19:46 (17.8s) 02:18:49 ~ 02:19:05 (14.2s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=612665167060641838456303519573845681516595451901075936837495812328875958

HTTP 头(TTFB) 2,763 ms 843 ms 2,046 ms 1,441 ms 2,359 ms

首文本(SSE delta) 2,768 ms 845 ms 2,047 ms 1,858 ms 2,360 ms

总耗时 16,318 ms 10,261 ms 17,631 ms 16,941 ms 13,443 ms

输入 3 164 164 164 164

输出 1,363 846 1,751 1,227 939

缓存创建

3,822

5m: 0 / 1h: 3,822

5m: 0 / 1h: 0

缓存读取 2,136 0 0 0 0

输出速度 100.6 tok/s 89.8 tok/s 112.4 tok/s 81.3 tok/s 84.7 tok/s

响应

Counting character by character in `s = "612665167060641838456303519573845681516595451901075936837495812328875958"`:

**Length:**
Breaking into groups of 10:
```
6126651670  (10)
6064183845  (20)
6303

**Parsing the string:**
`612665167060641838456303519573845681516595451901075936837495812328875958`

**Step 1 – Length:**
Counting all characters: **n = 72**

**Step 2 – Count 0s (positions 10,12,23,47

## Counting digits in s

**s** = `612665167060641838456303519573845681516595451901075936837495812328875958`

### Step-by-step indexing (0-based):

| idx | digit | idx | digit | idx | digit | idx | dig

```
s = "612665167060641838456303519573845681516595451901075936837495812328875958"
```

**Counting length (groups of 10):**
`6126651670` | `6064183845` | `6303519573` | `8456815165` | `9545190107` | `

```json
{"n":72,"c":[5,5,6]}
```

总计

总输入(含缓存) 103,011 138,255 138,823 138,300 138,208

总输出 1,655 1,132 2,039 1,523 1,219

总缓存创建

32,269

5m: 0 / 1h: 32,269

40,663

5m: 0 / 1h: 40,663

40,812

5m: 0 / 1h: 40,812

40,691

5m: 0 / 1h: 40,691

40,630

5m: 0 / 1h: 40,630

总缓存读取 70,724 97,413 97,832 97,430 97,399

输出速度 97.8 tok/s 91.2 tok/s 111.0 tok/s 81.0 tok/s 80.7 tok/s

总执行时间 46.4s 30.3s 51.7s 43.7s 40.8s

总墙钟时间

14m 53.0s

含步骤间等待

11m 3.9s

含步骤间等待

13m 42.0s

含步骤间等待

13m 32.9s

含步骤间等待

12m 55.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1671

Anthropic: Claude Sonnet 4.6

$0.1992

Anthropic: Claude Sonnet 4.6

$0.2135

Anthropic: Claude Sonnet 4.6

$0.2052

Anthropic: Claude Sonnet 4.6

$0.2004

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%) direct (100%) one-api (100%) direct (100%) one-api (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_013osR76X4oydYkQP3VS1JdT msg_01R6XB8SkFPv8CSbtQJvoLkJ msg_01EkqhnWEUFRzZHnBJHxzs4U msg_01FtEQMosGFHHjqZsFqDyqws msg_016q5PshFr3vREbqmKgeYFpt

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
R-kiro	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	—
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_011CcY1WXctBWtUZVYFDbq2n
O-Max	cf-ray: a136f0d0faee18ae-AMS server: cloudflare x-new-api-version: 0.12.11 x-oneapi-request-id: 202606291819393093227338268d9d6e6MyWYtl	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	a136f0d0faee18ae-AMS
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	—
M-	cf-ray: a136efa31a6afd4f-NRT server: cloudflare set-cookie: server_name_session=56fb82d09f6cfa691ede7bc922d16f50; Max-Age=86400; httponly; path=/ x-oneapi-request-id: 20260629181851578959766HsGQIrYJ	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-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a136efa31a6afd4f-NRT

协议指纹评分（机器自动）

总分 / 100 92 100 96 97 97

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 0.0 /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 7.5 /10 ×5 10.0 /10 ×5 4.1 /10 ×5 10.0 /10 ×5 10.0 /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 0.0 /10 ×4 10.0 /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 92 v3.25.0 基准（参考） 96 v3.25.0 97 v3.25.0 97 v3.25.0

雷达图例

共 20 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 档位思考量 tier_thinking_volume_match
20 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 6822 ms 3965 ms 7455 ms 6268 ms 5502 ms

输出速度 97.8 tok/s 91.2 tok/s 111.0 tok/s 81.0 tok/s 80.7 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 ⓘ

$0.1671

$0.1992

$0.2135

$0.2052

$0.2004

缓存读取占比 ⓘ 69% 70% 70% 70% 70%

揭晓

通道

服务商 ClaudeZZ ↗

通道R-kiro

官方基准

服务商 Aiide ↗

通道O-Max

服务商 AiTokensFlux ↗

通道O-Pro

服务商 LingxiCode ↗

通道M-

返回列表