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

2026/06/29 运行日期

claude-sonnet-4-6 官方基准

2 通道分数 < 70 与基线相似度

已揭晓揭晓进度

步骤 / 指标

FastCode

官方基准

LinkAPI

O-Max

0-0

O-Max

OmniaKey

O-Max

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 08:20:36 ~ 08:20:46 (8.4s) 08:20:38 ~ 08:20:48 (5.9s) 08:20:30 ~ 08:20:35 (4.7s) 08:20:32 ~ 08:20:37 (4.6s) 08:20:34 ~ 08:20:44 (6.6s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 6,248 ms 4,699 ms 3,662 ms 2,924 ms 5,150 ms

首文本(SSE delta) 6,249 ms 4,701 ms 3,663 ms 2,929 ms 5,151 ms

总耗时 6,858 ms 4,902 ms 3,744 ms 3,044 ms 5,292 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

32,318

5m: 0 / 1h: 32,318

31,364

5m: 31,364 / 1h: 0

32,506

5m: 32,506 / 1h: 0

32,318

5m: 0 / 1h: 32,318

缓存读取 0 0 0 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 08:23:08 ~ 08:23:18 (6.3s) 08:22:11 ~ 08:22:21 (3.7s) 08:23:56 ~ 08:24:01 (4.9s) 08:24:34 ~ 08:24:44 (6.7s) 08:23:46 ~ 08:23:56 (7.2s)

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。请按下列 3 行格式精确填空作答（label 不变，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) 4,076 ms 2,255 ms 3,250 ms 4,641 ms 5,510 ms

首文本(SSE delta) 4,077 ms 2,257 ms 3,252 ms 4,642 ms 5,511 ms

总耗时 4,846 ms 2,686 ms 3,644 ms 5,136 ms 5,973 ms

输入 3 3 3 3 3

输出 22 22 22 22 22

缓存创建

32,500

5m: 0 / 1h: 32,500

182

5m: 0 / 1h: 182

182

5m: 182 / 1h: 0

33,003

5m: 0 / 1h: 33,003

182

5m: 0 / 1h: 182

缓存读取 0 32,318 31,364 0 32,318

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 08:25:23 ~ 08:25:28 (3.7s) 08:25:02 ~ 08:25:22 (5.0s) 08:27:41 ~ 08:27:46 (4.0s) 08:26:23 ~ 08:27:38 (71.4s) 08:26:31 ~ 08:26:41 (5.4s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 2,333 ms 3,647 ms 2,796 ms 36,783 ms 4,032 ms

首文本(SSE delta) 2,335 ms 3,771 ms 2,914 ms - 4,181 ms

总耗时 2,335 ms 4,031 ms 3,018 ms 36,784 ms 4,206 ms

输入 3 3 3 3 3

输出 8 8 8 8 8

缓存创建

220

5m: 0 / 1h: 220

5m: 0 / 1h: 57

5m: 38 / 1h: 0

32,745

5m: 32,745 / 1h: 0

5m: 0 / 1h: 38

缓存读取 32,318 32,500 31,546 0 32,500

响应

2025-08

2025-08

2025-08

2025-08

2025-08

步骤 4: identity_free text

时间 08:27:12 ~ 08:27:22 (7.8s) 08:28:05 ~ 08:28:10 (4.5s) 08:28:48 ~ 08:28:58 (5.1s) 08:30:47 ~ 08:30:57 (7.7s) 08:29:44 ~ 08:29:54 (7.0s)

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) 3,762 ms 2,681 ms 2,672 ms 5,183 ms 4,801 ms

首文本(SSE delta) 3,763 ms 2,683 ms 3,078 ms 5,184 ms 4,802 ms

总耗时 5,521 ms 3,454 ms 4,112 ms 6,093 ms 5,755 ms

输入 3 3 3 3 3

输出 70 26 76 58 55

缓存创建

32,581

5m: 0 / 1h: 32,581

5m: 0 / 1h: 41

5m: 43 / 1h: 0

32,788

5m: 32,788 / 1h: 0

5m: 0 / 1h: 35

缓存读取 0 32,557 31,584 0 32,538

输出速度 39.8 tok/s - 73.5 tok/s 63.8 tok/s 57.7 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手。我擅长回答问题、编写代码、分析文本和协助各类任务，致力于提供安全、有帮助的智能对话服务。

我叫 Claude，是由 Anthropic 开发的 AI 助手。

我是 Claude，由 Anthropic 开发的 AI 助手。当前运行的是 Claude Sonnet 4.6 模型，作为 Claude Code CLI 工具在终端环境中为你提供编程和软件工程方面的帮助。

我是 Claude，由 Anthropic 开发的 AI 助手。当前运行的是 Claude Sonnet 4.6 模型，擅长代码、分析、写作等任务。

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。

步骤 5: knowledge_recall text

时间 08:28:31 ~ 08:28:41 (9.6s) 08:29:39 ~ 08:29:45 (4.2s) 08:32:16 ~ 08:32:31 (13.9s) 08:32:18 ~ 08:32:28 (6.9s) 08:32:20 ~ 08:32:30 (5.3s)

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) 1,942 ms 1,328 ms 3,877 ms 3,452 ms 2,369 ms

首文本(SSE delta) 1,944 ms 1,330 ms 5,119 ms 3,453 ms 2,370 ms

总耗时 8,454 ms 3,262 ms 12,749 ms 5,377 ms 4,027 ms

输入 3 3 3 3 3

输出 196 196 751 196 196

缓存创建

8,040

5m: 0 / 1h: 8,040

8,013

5m: 0 / 1h: 8,013

8,752

5m: 8,752 / 1h: 0

8,464

5m: 0 / 1h: 8,464

8,017

5m: 0 / 1h: 8,017

缓存读取 0 0 0 0 0

输出速度 30.1 tok/s 101.4 tok/s 98.4 tok/s 101.9 tok/s 118.3 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 08:31:13 ~ 08:31:38 (20.2s) 08:32:55 ~ 08:33:25 (26.6s) 08:35:12 ~ 08:35:27 (14.5s) 08:35:14 ~ 08:35:34 (17.5s) 08:34:41 ~ 08:34:56 (12.5s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=178012485349350958135901629876773460609087554997561596471872809207730241

HTTP 头(TTFB) 2,346 ms 1,451 ms 1,648 ms 1,936 ms 2,226 ms

首文本(SSE delta) 2,725 ms 1,453 ms 2,866 ms 1,938 ms 2,227 ms

总耗时 18,951 ms 25,544 ms 13,546 ms 15,943 ms 11,400 ms

输入 164 164 3 582 164

输出 1,348 1,378 1,191 1,281 848

缓存创建

5m: 0 / 1h: 0

1,924

5m: 1,924 / 1h: 0

5m: 0 / 1h: 0

缓存读取 0 0 0 0 0

输出速度 83.1 tok/s 57.2 tok/s 111.5 tok/s 91.5 tok/s 92.4 tok/s

响应

## Counting digits in s

**s** = `178012485349350958135901629876773460609087554997561596471872809207730241`

### Length
Grouped in 4s: `1780 1248 5349 3509 5813 5901 6298 7677 3460 6090 8755 4997 5615

**Step 1 — Length**

Splitting into groups of 10:
```
1780124853 | 4935095813 | 5901629876 | 7734606090 |
8755499756 | 1596471872 | 8092077302 | 41
```
7×10 + 2 = **72**

**Step 2 — Digit-by-digit sca

```json
{"n":72,"c":[9,10,9]}
```

{"n":72,"c":[9,10,9]}

```json
{"n":72,"c":[9,10,9]}
```

**Verification:**

```
s = 178012485349350958135901629876773460609087554997561596471872809207730241
```

Split into groups of 10 for counting:

| Chunk | 0s | 7s | 9

总计

总输入(含缓存) 138,156 138,165 136,815 140,103 138,125

总输出 1,649 1,635 2,053 1,570 1,134

总缓存创建

105,659

5m: 0 / 1h: 105,659

40,611

5m: 0 / 1h: 40,611

42,303

5m: 42,303 / 1h: 0

139,506

5m: 98,039 / 1h: 41,467

40,590

5m: 0 / 1h: 40,590

总缓存读取 32,318 97,375 94,494 0 97,356

输出速度 65.9 tok/s 60.5 tok/s 104.3 tok/s 91.2 tok/s 93.3 tok/s

总执行时间 55.8s 49.9s 47.0s 1m 54.7s 44.0s

总墙钟时间

11m 2.1s

含步骤间等待

12m 47.0s

含步骤间等待

14m 57.0s

含步骤间等待

15m 2.0s

含步骤间等待

14m 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 ⓘ

$0.4312

Anthropic: Claude Sonnet 4.6

$0.2066

Anthropic: Claude Sonnet 4.6

$0.1583

Anthropic: Claude Sonnet 4.6

$0.4106

Anthropic: Claude Sonnet 4.6

$0.1990

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 (75%) direct (100%) one-api (100%) direct (100%) one-api (100%)

上游 Anthropic 直连 (75%) Anthropic 直连 (100%) Anthropic 直连 (100%) Anthropic 直连 (100%) Anthropic 直连 (100%)

CDN - - - - cloudflare

ID 格式 msg_vrtx_01YEnbirD2PQTSqcCzxkSsmy msg_01JM3yxUr3erPjv8YGifCeSB msg_01Cj2xSCHwdhQnCnFAwjpXEQ msg_01LmH5X4iRs38x1f2ucBE7Md msg_01XzZaA1wqd5BWKjDNcSzDCQ

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
cc	server: nginx	id_format: msg_prefix extra_fields: ["stop_details"] has_service_tier: false has_inference_geo: false model_has_date_suffix: false usage_has_cache_fields: true	body:id:msg_prefixbody: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_011CcWcMmUtDA8awgdLudyvc
O-Max	server: nginx set-cookie: server_name_session=9a42ff809bdc7f49eaa4735d20f97eff; Max-Age=86400; httponly; path=/ x-new-api-version: v1.0.0-rc.15 x-oneapi-request-id: 202606290035132519616518268d9d6VPX1ZhW4	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-idbody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	—
O-Max	via: 1.1 Caddy server: uvicorn, nginx/1.24.0 (Ubuntu)	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:viabody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	4b4979f9a031f2bfca0ccb6c51e3a0ce
O-Max	cf-ray: a130d8d6ba8cb918-AMS server: cloudflare x-oneapi-request-id: 202606290034428518684758268d9d61UL2yWUT	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	a130d8d6ba8cb918-AMS

协议指纹评分（机器自动）

总分 / 100 70 100 86 68 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 0.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20 0.0 /10 ×20 10.0 /10 ×20

缓存 TTL 一致 10.0 /10 ×15 10.0 /10 ×15 0.0 /10 ×15 0.0 /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 5.3 /10 ×13 10.0 /10 ×13 10.0 /10 ×13 0.0 /10 ×13 10.0 /10 ×13

原生 msg-ID 0.0 /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 5.0 /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 0.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6

inference_geo 0.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

延迟基线 9.8 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 7.2 /10 ×5 9.2 /10 ×5

流式投递 10.0 /10 ×5 10.0 /10 ×5 7.9 /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 70 v3.25.0 基准（参考） 86 v3.25.0 68 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 7828 ms 7313 ms 6802 ms 12063 ms 6109 ms

输出速度 65.9 tok/s 60.5 tok/s 104.3 tok/s 91.2 tok/s 93.3 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 ⓘ

$0.4312

$0.2066

$0.1583

$0.4106

$0.1990

缓存读取占比 ⓘ 23% 70% 69% 0% 70%

揭晓

通道

服务商 FastCode ↗

通道cc

官方基准

服务商 LinkAPI ↗

通道O-Max

服务商 0-0 ↗

通道O-Max

服务商 OmniaKey ↗

通道O-Max

返回列表