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

2026/06/27 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

步骤 / 指标

FoxCode

官方基准

Aiide

O-Max

FastCode

AIMZ

M-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 17:13:09 ~ 17:13:19 (6.0s) 17:13:13 ~ 17:13:23 (7.2s) 17:13:11 ~ 17:13:16 (4.8s) 17:13:05 ~ 17:13:15 (5.4s) 17:13:07 ~ 17:13:17 (5.0s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 4,849 ms 6,201 ms 3,590 ms 2,938 ms 3,643 ms

首文本(SSE delta) 4,856 ms 6,203 ms 3,592 ms 2,939 ms 3,682 ms

总耗时 5,075 ms 6,205 ms 3,699 ms 2,939 ms 3,837 ms

输入 3 3 3 3 3

输出 5 5 5 5 5

缓存创建

32,318

5m: 32,318 / 1h: 0

32,318

5m: 0 / 1h: 32,318

32,457

5m: 0 / 1h: 32,457

32,318

5m: 0 / 1h: 32,318

31,985

5m: 0 / 1h: 31,985

缓存读取 0 0 0 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 17:17:12 ~ 17:17:22 (5.6s) 17:14:45 ~ 17:14:50 (4.9s) 17:15:08 ~ 17:15:18 (8.2s) 17:15:35 ~ 17:15:45 (6.7s) 17:16:05 ~ 17:16:15 (6.7s)

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) 4,279 ms 3,546 ms 4,197 ms 4,659 ms 4,378 ms

首文本(SSE delta) 4,304 ms 3,547 ms 4,198 ms 4,660 ms 4,689 ms

总耗时 4,751 ms 3,892 ms 4,552 ms 4,752 ms 5,565 ms

输入 3 3 3 3 3

输出 22 22 22 22 22

缓存创建

202

5m: 202 / 1h: 0

202

5m: 0 / 1h: 202

202

5m: 0 / 1h: 202

32,520

5m: 0 / 1h: 32,520

202

5m: 0 / 1h: 202

缓存读取 32,318 32,318 32,457 0 31,985

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-4-5

步骤 3: cutoff text

时间 17:20:10 ~ 17:20:20 (5.8s) 17:17:37 ~ 17:17:47 (6.5s) 17:17:15 ~ 17:17:25 (5.3s) 17:19:00 ~ 17:19:10 (5.9s) 17:20:12 ~ 17:20:22 (6.5s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) - 5,159 ms 3,562 ms 3,395 ms 2,809 ms

首文本(SSE delta) - 5,496 ms 3,675 ms 3,397 ms 2,910 ms

总耗时 - 5,507 ms 3,999 ms 4,246 ms 3,093 ms

输入 - 3 3 3 3

输出 - 8 8 8 8

缓存创建

5m: 0 / 1h: 57

32,577

5m: 0 / 1h: 32,577

5m: 0 / 1h: 55

缓存读取 - 32,520 32,659 0 32,187

响应

失败

评测失败

2025-08

2025-08

2025-08

2025-04

步骤 4: identity_free text

时间 - 17:18:52 ~ 17:19:02 (5.6s) 17:19:30 ~ 17:19:40 (5.3s) 17:22:40 ~ 17:22:50 (5.6s) 17:21:58 ~ 17:22:08 (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) - 4,136 ms 3,489 ms 2,962 ms 3,766 ms

首文本(SSE delta) - 4,137 ms 3,504 ms 2,963 ms 3,905 ms

总耗时 - 4,527 ms 4,342 ms 4,330 ms 4,182 ms

输入 - 3 3 3 3

输出 - 39 66 66 24

缓存创建

5m: 0 / 1h: 35

5m: 0 / 1h: 43

32,620

5m: 0 / 1h: 32,620

5m: 0 / 1h: 39

缓存读取 - 32,577 32,716 0 32,242

输出速度 - 100.0 tok/s 78.8 tok/s 48.3 tok/s -

响应

未执行

我是 Claude，由 Anthropic 开发的 AI 助手。当前版本为 claude-sonnet-4-6。

我是 Claude，由 Anthropic 开发的 AI 助手。我基于 claude-sonnet-4-6 模型，能够帮助用户完成写作、编程、分析和问答等各类任务。

我是 Claude，由 Anthropic 开发的 AI 助手。我基于 claude-sonnet-4-6 模型运行，能够帮助完成写作、分析、编程、问答等多种任务。

我是 Claude，由 Anthropic 打造的 AI 助手。

步骤 5: knowledge_recall text

时间 - 17:22:03 ~ 17:22:13 (6.2s) 17:23:11 ~ 17:23:21 (6.3s) 17:24:47 ~ 17:24:57 (5.7s) 17:24:20 ~ 17:24:30 (5.6s)

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) - 3,040 ms 3,315 ms 2,444 ms 2,060 ms

首文本(SSE delta) - 3,042 ms 3,383 ms 2,445 ms 4,448 ms

总耗时 - 5,201 ms 5,149 ms 4,493 ms 4,451 ms

输入 - 3 3 3 3

输出 - 204 205 205 198

缓存创建

8,040

5m: 0 / 1h: 8,040

8,075

5m: 0 / 1h: 8,075

8,075

5m: 0 / 1h: 8,075

7,694

5m: 0 / 1h: 7,694

缓存读取 - 0 0 0 0

输出速度 - 94.5 tok/s 116.1 tok/s 100.1 tok/s burst

响应

未执行

{"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 - 17:25:06 ~ 17:25:31 (20.6s) 17:25:28 ~ 17:25:48 (19.1s) 17:26:37 ~ 17:27:07 (29.3s) 17:27:50 ~ 17:28:10 (19.4s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=586611209375395065150044550713752479160730097012747434737085957650659543

HTTP 头(TTFB) - 2,210 ms 2,381 ms 1,602 ms 2,363 ms

首文本(SSE delta) - 2,212 ms 2,382 ms 1,959 ms 2,687 ms

总耗时 - 19,642 ms 17,673 ms 28,152 ms 18,321 ms

输入 - 164 164 164 110

输出 - 1,642 1,510 2,389 1,488

缓存创建

5m: 0 / 1h: 0

缓存读取 - 0 0 0 0

输出速度 - 94.2 tok/s 98.8 tok/s 91.2 tok/s 95.2 tok/s

响应

未执行

## Counting digits in s

**s** = `586611209375395065150044550713752479160730097012747434737085957650659543`

### Length

Splitting into groups of 10:
```
5866120937  (10)
5395065150  (10)
0445507137

```
s = 586611209375395065150044550713752479160730097012747434737085957650659543
```

**Length (n):**
Splitting into groups of 10:
`5866112093` `7539506515` `0044550713` `7524791607` `3009701274` `743

## Counting digits in s

**s** = `586611209375395065150044550713752479160730097012747434737085957650659543`

### Length
Splitting into groups of 10:
```
5866112093 | 7539506515 | 0044550713 | 75247916

```
s = "586611209375395065150044550713752479160730097012747434737085957650659543"
```

Splitting into groups of 10 for careful counting:

| # | Group | 0s | 7s | 9s |
|---|-------|----|----|----|
| 1

总计

总输入(含缓存)

64,844

仅前 3 步

138,246

138,845

138,289

136,514

总输出

仅前 3 步

1,920

1,816

2,695

1,745

总缓存创建

32,520

5m: 32,520 / 1h: 0

仅前 3 步

40,652

5m: 0 / 1h: 40,652

40,834

5m: 0 / 1h: 40,834

138,110

5m: 0 / 1h: 138,110

39,975

5m: 0 / 1h: 39,975

总缓存读取

32,318

仅前 3 步

97,415

97,832

96,414

输出速度

仅前 3 步

94.3 tok/s

99.5 tok/s

89.8 tok/s

95.2 tok/s

总执行时间

17.4s

仅前 3 步

50.9s

49.0s

58.6s

48.4s

总墙钟时间

7m 10.4s

含步骤间等待

仅前 3 步

12m 18.0s

含步骤间等待

12m 37.0s

含步骤间等待

14m 1.7s

含步骤间等待

15m 3.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.0863

Anthropic: Claude Sonnet 4.6 (部分)

$0.2110

Anthropic: Claude Sonnet 4.6

$0.2103

Anthropic: Claude Sonnet 4.6

$0.5589

Anthropic: Claude Sonnet 4.6

$0.2054

Anthropic: Claude Sonnet 4.6

通道指纹

来源步骤 identity digit_count digit_count digit_count digit_count

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

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

CDN cloudflare - cloudflare - cloudflare

ID 格式 msg_0171P7hzrri5cXVfqDykCx5Y msg_015KsRUcfmaPfbxDMxVrFuYn msg_016ECCLb6J3vFXkXKXij3vrx msg_vrtx_01DCUst5j3Hq5NHKtfhRctCx msg_01MqicvkspdQm1tXtuTkdtpE

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	cf-ray: a1235b807b55734b-NRT server: cloudflare set-cookie: _cfuvid=ZA6KBhhg60utDVkYdIxH8Gso9IhNSMyrz0m0duZhArs-1782551834.3307614-1.0.1.1-RV_OttKdbgs8lGz2Cdvsr142A75uzXC.9jGBeAj4PH8; HttpOnly; SameSite=None; Secure; Path=/; Domain=api.anthropic.com	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	req_011CcTWipCyFo7NmddeNiSwp a1235b807b55734b-NRT
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_011CcTXKnoMtZnFrYJPapgkb
O-Max	cf-ray: a123679f88d4bb09-AMS server: cloudflare x-new-api-version: 0.12.11 x-oneapi-request-id: 202606270925303820943768268d9d6UKlpDYgd	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	a123679f88d4bb09-AMS
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	—
M-Max	cf-ray: a1236b148cb1131e-AMS server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 202606270927517868061838268d9d62wcJRgDd	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	a1236b148cb1131e-AMS

协议指纹评分（机器自动）

总分 / 100 100 98 66 86

基线状态已对照基线已对照基线已对照基线已对照基线

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 0.0 /10 ×13 10.0 /10 ×13

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

缓存命中比 — 10.0 /10 ×20 10.0 /10 ×20 0.0 /10 ×20 10.0 /10 ×20

身份(结构) — 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 6.7 /10 ×7

知识截止 — 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 5.0 /10 ×7

身份(自由) — 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7 10.0 /10 ×7

原生 msg-ID — 10.0 /10 ×8 10.0 /10 ×8 0.0 /10 ×8 10.0 /10 ×8

Req-ID 透传 — 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

service_tier — 10.0 /10 ×6 10.0 /10 ×6 0.0 /10 ×6 10.0 /10 ×6

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

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

系统提示纯净 — 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 5.0 /10 ×8

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 10.0 /10 ×5 10.0 /10 ×5 0.0 /10 ×5

综合结论

与基线相似度执行失败，不评分基准（参考） 98 v3.25.0 66 v3.25.0 86 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 4913 ms 7496 ms 6569 ms 8152 ms 6575 ms

输出速度 - 94.3 tok/s 99.5 tok/s 89.8 tok/s 95.2 tok/s

完成步骤 2/6 成功 · 1 失败 · 3 跳过 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 ⓘ

$0.0863 (部分)

$0.2110

$0.2103

$0.5589

$0.2054

缓存读取占比 ⓘ 50% (部分) 70% 70% 0% 71%

揭晓

通道

服务商 FoxCode ↗

通道cc

官方基准

服务商 Aiide ↗

通道O-Max

服务商 FastCode ↗

通道cc

服务商 AIMZ ↗

通道M-Max

返回列表