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

2026/06/24 运行日期

claude-sonnet-4-6 官方基准

1 任务失败执行不完整

已揭晓揭晓进度

步骤 / 指标

ByteCatCode

O-Max

ClaudeCN

O-Max

DawAPI

官方基准

0-0

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 23:37:46 ~ 23:37:56 (7.1s) 23:37:50 ~ 23:38:00 (7.7s) 23:37:44 ~ 23:37:49 (4.1s) 23:37:52 ~ 23:37:57 (3.5s) 23:37:48 ~ 23:37:58 (7.5s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 4,603 ms 5,985 ms 2,868 ms 2,098 ms 3,945 ms

首文本(SSE delta) - 5,989 ms 2,869 ms 2,101 ms 4,218 ms

总耗时 5,559 ms 5,989 ms 3,126 ms 2,239 ms 4,219 ms

输入 3 3 3 3 3

输出 5 5 5 5 5

缓存创建

32,346

5m: 0 / 1h: 32,346

32,319

5m: 0 / 1h: 32,319

32,318

5m: 0 / 1h: 32,318

32,318

5m: 0 / 1h: 32,318

32,805

5m: 0 / 1h: 32,805

缓存读取 0 0 0 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 23:39:21 ~ 23:39:31 (6.3s) 23:41:05 ~ 23:41:15 (6.0s) 23:41:39 ~ 23:41:49 (6.5s) 23:39:10 ~ 23:39:15 (3.5s) 23:41:00 ~ 23:41:05 (3.7s)

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) 4,214 ms 4,352 ms 5,392 ms 2,207 ms -

首文本(SSE delta) - 4,379 ms 5,393 ms 2,209 ms -

总耗时 5,120 ms 4,713 ms 5,617 ms 2,521 ms -

输入 3 3 3 3 -

输出 22 22 22 22 -

缓存创建

32,548

5m: 0 / 1h: 32,548

202

5m: 0 / 1h: 202

32,518

5m: 0 / 1h: 32,518

202

5m: 0 / 1h: 202

缓存读取 0 32,319 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

失败

评测失败

步骤 3: cutoff text

时间 23:40:53 ~ 23:41:03 (5.1s) 23:42:31 ~ 23:42:41 (6.8s) 23:44:33 ~ 23:44:43 (9.2s) 23:42:15 ~ 23:42:20 (3.8s) -

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 3,534 ms 4,428 ms 8,186 ms 2,341 ms -

首文本(SSE delta) - 4,429 ms 8,309 ms 2,343 ms -

总耗时 3,987 ms 4,931 ms 8,402 ms 2,590 ms -

输入 3 3 3 3 -

输出 8 8 8 8 -

缓存创建

5m: 0 / 1h: 38

5m: 0 / 1h: 55

5m: 0 / 1h: 38

5m: 0 / 1h: 55

缓存读取 32,548 32,521 32,518 32,520 -

响应

2025-08

2025-08

2025-08

2025-08

未执行

步骤 4: identity_free text

时间 23:44:09 ~ 23:44:19 (5.7s) 23:44:20 ~ 23:44:30 (7.6s) 23:48:22 ~ 23:48:33 (9.3s) 23:44:57 ~ 23:45:07 (5.3s) -

提示词

查看本步 Prompt

用中文回答：你是谁？由谁开发？50 字以内。

HTTP 头(TTFB) 3,287 ms 5,304 ms 6,113 ms 2,831 ms -

首文本(SSE delta) - 5,305 ms 6,501 ms 2,833 ms -

总耗时 4,617 ms 6,549 ms 8,423 ms 4,273 ms -

输入 3 3 3 3 -

输出 55 64 66 62 -

缓存创建

5m: 0 / 1h: 35

5m: 0 / 1h: 43

缓存读取 32,586 32,576 32,556 32,575 -

输出速度 41.4 tok/s 51.4 tok/s 34.3 tok/s 43.1 tok/s -

响应

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。我基于 claude-sonnet-4-6 模型，能够帮助完成写作、编程、分析、问答等多种任务。

我是 Claude，由 Anthropic 开发的 AI 助手。当前运行的是 Claude Code，专为软件工程任务设计的命令行工具，基于 claude-sonnet-4-6 模型。

我是 Claude，由 Anthropic 开发的 AI 助手。我基于 claude-sonnet-4-6 模型，能够帮助你完成编程、写作、分析等各类任务。

未执行

步骤 5: knowledge_recall text

时间 23:46:54 ~ 23:47:04 (6.9s) 23:47:24 ~ 23:47:34 (7.5s) 23:52:30 ~ 23:52:40 (6.0s) 23:48:04 ~ 23:48:14 (5.0s) -

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) 2,757 ms 2,698 ms 2,928 ms 1,634 ms -

首文本(SSE delta) - 3,866 ms 2,930 ms 1,635 ms -

总耗时 5,033 ms 6,014 ms 4,805 ms 3,989 ms -

输入 3 3 3 3 -

输出 205 205 205 204 -

缓存创建

8,065

5m: 0 / 1h: 8,065

8,072

5m: 0 / 1h: 8,072

8,054

5m: 0 / 1h: 8,054

8,069

5m: 0 / 1h: 8,069

缓存读取 0 0 0 0 -

输出速度 90.1 tok/s 95.4 tok/s 109.3 tok/s 86.7 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

未执行

步骤 6: digit_count text

时间 23:49:18 ~ 23:49:43 (22.6s) 23:51:30 ~ 23:51:45 (14.9s) 23:55:07 ~ 23:55:22 (10.6s) 23:50:34 ~ 23:50:49 (12.0s) -

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=736231113237312702901842103444277517691266324220657656323365986631817788

HTTP 头(TTFB) 4,407 ms 4,421 ms 1,683 ms 1,732 ms -

首文本(SSE delta) - 4,423 ms 8,116 ms 2,121 ms -

总耗时 21,590 ms 13,012 ms 9,552 ms 11,000 ms -

输入 3 164 193 164 -

输出 1,426 625 735 838 -

缓存创建

30,659

5m: 0 / 1h: 30,659

5m: 0 / 1h: 0

缓存读取 0 0 0 0 -

输出速度 83.0 tok/s 72.8 tok/s burst 94.4 tok/s -

响应

Counting by indexing each character (1-based):

```
7(1)3(2)6(3)2(4)3(5)1(6)1(7)1(8)3(9)2(10)
3(11)7(12)3(13)1(14)2(15)7(16)0(17)2(18)9(19)0(20)
1(21)8(22)4(23)2(24)1(25)0(26)3(27)4(28)4(29)4(30)
2(31

```json
{"n":72,"c":[4,9,3]}
```

```json
{"n":72,"c":[4,9,3]}
```

**Verification walkthrough:**

`736231113237312702901842103444277517691266324220657656323365986631817788`

- **Length:** 72 digits
- **0s** (positions 17,20,26,48): *

```json
{"n":72,"c":[4,9,3]}
```

**Verification walkthrough:**

```
s = 736231113237312702901842103444277517691266324220657656323365986631817788
     
Length: 72 characters

Digit 0: positions 17,20,

未执行

总计

总输入(含缓存)

168,843

138,286

138,253

138,279

32,808

仅前 2 步

总输出

1,721

929

1,041

1,139

仅前 2 步

总缓存创建

103,691

5m: 0 / 1h: 103,691

40,691

5m: 0 / 1h: 40,691

72,971

5m: 0 / 1h: 72,971

40,687

5m: 0 / 1h: 40,687

32,805

5m: 0 / 1h: 32,805

仅前 2 步

总缓存读取

65,134

97,416

65,074

97,413

仅前 2 步

输出速度

81.1 tok/s

74.6 tok/s

71.4 tok/s

87.1 tok/s

仅前 2 步

总执行时间

53.7s

50.4s

45.7s

33.1s

11.2s

仅前 2 步

总墙钟时间

11m 57.0s

含步骤间等待

13m 55.3s

含步骤间等待

17m 38.1s

含步骤间等待

12m 56.9s

含步骤间等待

3m 17.2s

含步骤间等待

仅前 2 步

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.4343

Anthropic: Claude Sonnet 4.6

$0.1963

Anthropic: Claude Sonnet 4.6

$0.3094

Anthropic: Claude Sonnet 4.6

$0.1994

Anthropic: Claude Sonnet 4.6

$0.1231

Anthropic: Claude Sonnet 4.6 (部分)

通道指纹

来源步骤 digit_count digit_count digit_count digit_count ping

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

ID 格式 msg_01Eo5EuzNhRWLEfmYsLRpa3Z msg_016GK4hoBtxabg7RzDieeDEx msg_01WcTKk46pQasBrNqPwKjuR4 msg_01GyYjgpoVCSDJM4uGjtXfdq msg_01UeRwWwAehFZN6n8D2jVvcp

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	server: nginx/1.29.8 x-oneapi-request-id: 202606241549193504502268268d9d6DdqkeX2R	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-Max	cf-ray: a10ce4fa7b735671-AMS server: cloudflare x-oneapi-request-id: 202606242351327962684352mpqfjaJ	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	a10ce4fa7b735671-AMS
cc	server: cdn x-new-api-version: v1.0.0-rc.14 x-oneapi-request-id: 202606241555084624365548268d9d61K48allH	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	—
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_011CcNMHpqAP8sE8mbMtxfAG
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	9a3fbd67f65bc3d28dfee60c131a54dd

协议指纹评分（机器自动）

总分 / 100 90 97 90 99

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 5.6 /10 ×20 10.0 /10 ×20 5.6 /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 —

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 7.7 /10 ×13 10.0 /10 ×13 7.7 /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 —

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 —

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 8.2 /10 ×5 10.0 /10 ×5 —

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

流式投递未计入 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 —

综合结论

与基线相似度 90 v3.24.1 97 v3.24.1 90 v3.24.1 基准（参考）执行失败，不评分

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 自由身份 identity_free_clean
5 停止原因 stop_reason_present
6 系统提示纯净 system_prompt_clean
7 Service Tier service_tier_present
8 缓存命中比 cache_hit_ratio_match
9 缓存 TTL 一致 cache_ttl_consistency
10 推理地区 inference_geo_present
11 缓存连续 cache_continuity_intra
12 知识召回 knowledge_recall_match
13 延迟基线 latency_baseline_match
14 原生 msg-ID anthropic_msg_id_format
15 Sliding 缓存 cache_sliding_correctness
16 结构化身份 identity_structured_match
17 信封自报一致 envelope_self_report_match
18 中文遵循 instruction_following_lang
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 7651 ms 6868 ms 6654 ms 4435 ms 4219 ms

输出速度 81.1 tok/s 74.6 tok/s 71.4 tok/s 87.1 tok/s -

完成步骤 6/6 成功 6/6 成功 6/6 成功 6/6 成功 1/6 成功 · 1 失败 · 4 跳过

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.4343

$0.1963

$0.3094

$0.1994

$0.1231 (部分)

缓存读取占比 ⓘ 39% 70% 47% 70% 0% (部分)

揭晓

通道

服务商 ByteCatCode ↗

通道O-Max

服务商 ClaudeCN ↗

通道O-Max

服务商 DawAPI ↗

通道cc

官方基准

服务商 0-0 ↗

通道O-Max

返回列表