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。

4 通道 · 6 步骤

2026/06/22 运行日期

claude-sonnet-4-6 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

Aiberm / cc
```
评测失败
```

步骤 / 指标

ClaudeCN

O-Max

TopRouterCN

O-Max

官方基准

WorldBase.ai

M-

基本信息

服务名 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

客户端 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 10:05:07 ~ 10:05:22 (14.5s) 10:05:01 ~ 10:05:06 (4.0s) 10:05:09 ~ 10:05:14 (4.2s) 10:05:05 ~ 10:05:15 (7.0s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 12,685 ms 3,124 ms 2,794 ms 4,841 ms

首文本(SSE delta) 12,687 ms 3,125 ms 2,796 ms 4,849 ms

总耗时 13,011 ms 3,243 ms 2,944 ms 5,297 ms

输入 3 1,467 3 23

输出 5 5 5 24

缓存创建

32,318

5m: 0 / 1h: 32,318

30,854

5m: 0 / 1h: 30,854

32,318

5m: 0 / 1h: 32,318

32,993

5m: 32,993 / 1h: 0

缓存读取 0 0 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 10:08:52 ~ 10:09:02 (5.7s) 10:08:36 ~ 10:08:41 (4.4s) 10:07:41 ~ 10:07:46 (3.4s) 10:09:02 ~ 10:09:12 (6.8s)

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) 4,290 ms 3,475 ms 2,028 ms 3,710 ms

首文本(SSE delta) 4,291 ms 3,476 ms 2,029 ms 3,712 ms

总耗时 4,731 ms 3,558 ms 2,417 ms 4,462 ms

输入 3 3 3 24

输出 22 22 22 48

缓存创建

189

5m: 0 / 1h: 189

1,653

5m: 0 / 1h: 1,653

189

5m: 0 / 1h: 189

33,521

5m: 33,521 / 1h: 0

缓存读取 32,318 30,854 32,318 0

输出速度 - - - 64.0 tok/s

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

步骤 3: cutoff text

时间 10:11:21 ~ 10:11:31 (7.2s) 10:11:15 ~ 10:11:20 (4.9s) 10:11:17 ~ 10:11:22 (4.5s) 10:11:34 ~ 10:11:44 (7.5s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 5,667 ms 4,072 ms 3,035 ms 5,705 ms

首文本(SSE delta) 5,668 ms 4,073 ms 3,037 ms 5,707 ms

总耗时 5,949 ms 4,158 ms 3,276 ms 6,070 ms

输入 3 3 3 24

输出 8 8 8 37

缓存创建

5m: 0 / 1h: 38

5m: 0 / 1h: 57

385

5m: 385 / 1h: 0

缓存读取 32,507 32,507 32,507 33,521

输出速度 - - - 101.9 tok/s

响应

2025-08

2025-08

2025-08

2025-08

步骤 4: identity_free text

时间 10:13:09 ~ 10:13:19 (7.0s) 10:15:20 ~ 10:15:30 (5.8s) 10:12:24 ~ 10:12:34 (7.4s) 10:13:13 ~ 10:13:23 (6.9s)

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) 4,326 ms 4,886 ms 5,887 ms 4,208 ms

首文本(SSE delta) 4,327 ms 4,887 ms 5,888 ms 4,209 ms

总耗时 5,205 ms 5,001 ms 6,359 ms 5,677 ms

输入 3 3 3 71

输出 44 54 40 67

缓存创建

5m: 0 / 1h: 39

5m: 0 / 1h: 35

5m: 0 / 1h: 41

326

5m: 326 / 1h: 0

缓存读取 32,545 32,545 32,564 33,906

输出速度 50.1 tok/s - 84.9 tok/s 45.6 tok/s

响应

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

我是Claude，由Anthropic开发的AI助手。我基于claude-sonnet-4-6模型，致力于提供安全、有帮助的对话和任务协助。

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

我叫 Claude，是 Anthropic 开发的 AI 助手，当前使用的模型版本为 claude-sonnet-4-6。

步骤 5: knowledge_recall text

时间 10:16:53 ~ 10:17:03 (6.5s) 10:17:24 ~ 10:17:34 (7.7s) 10:14:21 ~ 10:14:31 (5.1s) 10:17:06 ~ 10:17:36 (25.8s)

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) 2,803 ms 6,577 ms 1,665 ms 2,564 ms

首文本(SSE delta) 2,804 ms 6,578 ms 1,667 ms 2,570 ms

总耗时 5,297 ms 6,633 ms 3,985 ms 24,679 ms

输入 3 3 3 25

输出 205 204 205 1,857

缓存创建

8,017

5m: 0 / 1h: 8,017

8,023

5m: 0 / 1h: 8,023

8,034

5m: 0 / 1h: 8,034

9,644

5m: 9,644 / 1h: 0

缓存读取 0 0 0 0

输出速度 82.2 tok/s burst 88.4 tok/s 84.0 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

步骤 6: digit_count text

时间 10:20:36 ~ 10:20:56 (16.0s) 10:21:20 ~ 10:21:45 (20.1s) 10:16:18 ~ 10:16:38 (17.3s) 10:20:13 ~ 10:20:33 (15.6s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=826602983292267861603681912969259897618943645711315916242089710988367581

HTTP 头(TTFB) 2,341 ms 6,738 ms 1,383 ms 4,125 ms

首文本(SSE delta) 2,342 ms 6,739 ms 1,781 ms 4,128 ms

总耗时 13,776 ms 19,366 ms 16,358 ms 14,502 ms

输入 164 164 164 473

输出 920 1,316 1,307 502

缓存创建

5m: 0 / 1h: 0

缓存读取 0 0 0 0

输出速度 80.5 tok/s 104.2 tok/s 89.7 tok/s 48.4 tok/s

响应

```json
{"n":72,"c":[4,5,11]}
```

**Verification:**

| Digit | Positions (1-indexed) | Count |
|---|---|---|
| **0** | 5, 20, 58, 63 | **4** |
| **7** | 15, 36, 46, 61, 69 | **5** |
| **9** | 7, 11,

## Counting digits in s

**s** = `826602983292267861603681912969259897618943645711315916242089710988367581`

### Step-by-step parse (indexed):
| # | d | # | d | # | d | # | d | # | d | # | d |
|---|--

## Counting digits in s

**s** = `826602983292267861603681912969259897618943645711315916242089710988367581`

### Length
Grouping in blocks of 4:
`8266 0298 3292 2678 6160 3681 9129 6925 9897 6189 4364

{"n":72,"c":[4,5,11]}

总计

总输入(含缓存) 138,150 138,152 138,207 144,936

总输出 1,204 1,609 1,587 2,535

总缓存创建

40,601

5m: 0 / 1h: 40,601

40,603

5m: 0 / 1h: 40,603

40,639

5m: 0 / 1h: 40,639

76,869

5m: 76,869 / 1h: 0

总缓存读取 97,370 95,906 97,389 67,427

输出速度 79.0 tok/s 104.2 tok/s 89.4 tok/s 71.6 tok/s

总执行时间 56.9s 46.9s 42.0s 1m 9.7s

总墙钟时间

15m 49.0s

含步骤间等待

16m 44.1s

含步骤间等待

11m 29.2s

含步骤间等待

15m 28.1s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2001

Anthropic: Claude Sonnet 4.6

$0.2101

Anthropic: Claude Sonnet 4.6

$0.2060

Anthropic: Claude Sonnet 4.6

$0.2403

Anthropic: Claude Sonnet 4.6

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

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

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

CDN cloudflare - - cloudflare

ID 格式 msg_01MEXYCWhL3wDWHaMkFb6Ezt msg_01C9NPY35DrXvoRDdD7kkkKd msg_019EJoMtbS8uvnHUjnoVGhaR msg_3e3f7d194efd439ea10f9d11ccfc6e63

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-Max	cf-ray: a0f7c667ddd2e00a-AMS server: cloudflare x-oneapi-request-id: 20260622102039595377612TE41LKBi	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	a0f7c667ddd2e00a-AMS
O-Max	server: nginx/1.27.5	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	tr-ew7y4grd-tlfecbjbfy-009885
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_011CcHVadbNEHp74V9ZjH8SJ
M-	cf-ray: a0f7c5d13f71adc4-AMS server: cloudflare	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:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	e130f689-3df1-4046-bb67-b20edce71ce5 a0f7c5d13f71adc4-AMS

协议指纹评分（机器自动）

总分 / 100 98 94 100 77

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 10.0 /10 ×20 5.4 /10 ×20

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

模型匹配 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

sliding 5m 10.0 /10 ×13 10.0 /10 ×13 10.0 /10 ×13 7.7 /10 ×13

原生 msg-ID 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 5.0 /10 ×8

身份(结构) 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

身份(自由) 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

inference_geo 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

流式投递 10.0 /10 ×5 0.0 /10 ×5 10.0 /10 ×5 9.9 /10 ×5

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

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

SDK 一致 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 98 v3.24.0 94 v3.24.0 基准（参考） 77 v3.24.0

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 7995 ms 6993 ms 5890 ms 10115 ms

输出速度 79.0 tok/s 104.2 tok/s 89.4 tok/s 71.6 tok/s

完成步骤 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 ⓘ

$0.2001

$0.2101

$0.2060

$0.2403

缓存读取占比 ⓘ 70% 69% 70% 47%

揭晓

通道

服务商 ClaudeCN ↗

通道O-Max

服务商 TopRouterCN ↗

通道O-Max

官方基准

服务商 WorldBase.ai ↗

通道M-

返回列表