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

2026/06/18 运行日期

claude-sonnet-4-6 官方基准

2 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

InfAI / R-Kiro
```
评测失败
```

步骤 / 指标

0-0

O-Max

FoxCode

R-ultra

DawAPI

官方基准

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 08:47:03 ~ 08:47:13 (6.8s) 08:47:07 ~ 08:47:12 (4.0s) 08:47:05 ~ 08:47:15 (6.9s) 08:47:09 ~ 08:47:19 (5.7s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 4,526 ms 2,466 ms 5,262 ms 3,837 ms

首文本(SSE delta) 4,527 ms 2,468 ms 5,264 ms 3,838 ms

总耗时 4,937 ms 3,036 ms 6,144 ms 3,991 ms

输入 3 48 3 3

输出 5 15 5 5

缓存创建

32,805

5m: 32,805 / 1h: 0

32,968

5m: 32,968 / 1h: 0

27,260

5m: 27,260 / 1h: 0

32,318

5m: 0 / 1h: 32,318

缓存读取 0 0 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 08:48:59 ~ 08:49:09 (6.3s) 08:50:36 ~ 08:50:46 (5.1s) 08:50:41 ~ 08:50:51 (5.3s) 08:51:17 ~ 08:51:22 (4.9s)

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。自报身份，输出严格遵守以下 3 行格式（不允许 markdown、代码块、额外文本）：
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) 4,404 ms 3,535 ms 4,250 ms 3,430 ms

首文本(SSE delta) 4,406 ms 3,537 ms 4,251 ms 3,431 ms

总耗时 4,494 ms 4,099 ms 4,508 ms 3,873 ms

输入 3 95 3 3

输出 22 81 22 22

缓存创建

7,635

5m: 7,635 / 1h: 0

33,436

5m: 33,436 / 1h: 0

192

5m: 192 / 1h: 0

192

5m: 0 / 1h: 192

缓存读取 25,362 0 27,260 32,318

输出速度 - 144.1 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-4-5

vendor: anthropic
brand: claude
model: claude-sonnet-4-6

步骤 3: cutoff text

时间 08:51:43 ~ 08:51:53 (6.5s) 08:53:24 ~ 08:53:29 (4.5s) 08:54:15 ~ 08:54:20 (4.6s) 08:52:51 ~ 08:52:56 (4.5s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 4,558 ms 3,211 ms 3,619 ms 3,204 ms

首文本(SSE delta) 4,667 ms 3,212 ms 3,751 ms 3,332 ms

总耗时 4,753 ms 3,542 ms 3,833 ms 3,411 ms

输入 3 49 3 3

输出 8 50 8 8

缓存创建

7,672

5m: 7,672 / 1h: 0

33,894

5m: 33,894 / 1h: 0

5m: 57 / 1h: 0

5m: 0 / 1h: 38

缓存读取 25,362 0 27,452 32,510

响应

2025-08

2025-08

2025-04

2025-08

步骤 4: identity_free text

时间 08:53:38 ~ 08:53:48 (6.7s) 08:55:41 ~ 08:55:51 (7.2s) 08:56:15 ~ 08:56:25 (6.3s) 08:55:47 ~ 08:55:52 (4.0s)

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) 4,741 ms 4,044 ms 4,352 ms 2,353 ms

首文本(SSE delta) 4,742 ms 4,045 ms 4,627 ms 2,354 ms

总耗时 5,198 ms 6,338 ms 5,462 ms 2,963 ms

输入 3 100 3 3

输出 44 166 24 43

缓存创建

7,713

5m: 7,713 / 1h: 0

340

5m: 340 / 1h: 0

5m: 35 / 1h: 0

5m: 0 / 1h: 41

缓存读取 25,362 33,894 27,509 32,548

输出速度 96.5 tok/s 72.4 tok/s - 70.6 tok/s

响应

我是 Claude，由 Anthropic 公司打造的 AI 助手。目前运行的版本是 Claude Sonnet 4.6。

我是 Kiro，一个由 Anthropic 技术驱动的 AI 智能助手，基于 Claude 模型构建，专注于协助软件开发及各类专业任务。

我是 Claude，由 Anthropic 开发的 AI 助手。

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

步骤 5: knowledge_recall text

时间 08:56:04 ~ 08:56:24 (15.2s) 08:59:18 ~ 08:59:43 (22.7s) 08:59:20 ~ 08:59:30 (5.1s) 08:59:33 ~ 08:59:38 (4.5s)

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) 3,204 ms 3,620 ms 2,404 ms 1,398 ms

首文本(SSE delta) 3,558 ms 3,689 ms 2,405 ms 1,400 ms

总耗时 13,703 ms 21,882 ms 4,232 ms 3,326 ms

输入 3 54 3 3

输出 720 1,572 198 204

缓存创建

8,437

5m: 8,437 / 1h: 0

9,754

5m: 9,754 / 1h: 0

3,160

5m: 3,160 / 1h: 0

8,021

5m: 0 / 1h: 8,021

缓存读取 0 0 0 0

输出速度 71.0 tok/s 86.4 tok/s 108.4 tok/s 105.9 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 08:58:39 ~ 08:58:59 (19.6s) 09:00:54 ~ 09:01:19 (24.8s) 09:02:08 ~ 09:02:28 (14.9s) 09:03:26 ~ 09:03:46 (16.3s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=967423340186027718719769878367441991743233159703000079141353779806548658

HTTP 头(TTFB) 2,979 ms 2,142 ms 1,105 ms 1,861 ms

首文本(SSE delta) 2,980 ms 2,143 ms 10,987 ms 2,254 ms

总耗时 17,855 ms 23,909 ms 14,190 ms 15,357 ms

输入 581 93 1,001 164

输出 1,078 2,601 1,395 927

缓存创建

5m: 0 / 1h: 0

115

5m: 115 / 1h: 0

5m: 0 / 1h: 0

缓存读取 0 0 0 0

输出速度 72.5 tok/s 119.5 tok/s burst 70.7 tok/s

响应

{"n":72,"c":[8,12,8]}

{"n":72,"c":[8,12,8]}

Here's the exact count by indexing each character:

```
s = "967423340186027718719769878367441991743233159703000079141353779806548658"

Index by 10s:
[00-09]: 9 6 7 4 2 3 3 4 0 1
[10-19]: 8 6 0 2 7 7

```json
{"n":72,"c":[8,12,7]}
```

总计

总输入(含缓存) 140,944 144,840 113,941 138,165

总输出 1,877 4,485 1,652 1,209

总缓存创建

64,262

5m: 64,262 / 1h: 0

110,507

5m: 110,507 / 1h: 0

30,704

5m: 30,704 / 1h: 0

40,610

5m: 0 / 1h: 40,610

总缓存读取 76,086 33,894 82,221 97,376

输出速度 72.3 tok/s 103.2 tok/s 108.4 tok/s 75.1 tok/s

总执行时间 1m 1.1s 1m 8.3s 43.1s 39.9s

总墙钟时间

11m 56.0s

含步骤间等待

14m 12.1s

含步骤间等待

15m 23.0s

含步骤间等待

16m 37.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2034

Anthropic: Claude Sonnet 4.6

$0.3378

Anthropic: Claude Sonnet 4.6

$0.1245

Anthropic: Claude Sonnet 4.6

$0.2002

Anthropic: Claude Sonnet 4.6

通道指纹

来源步骤 digit_count digit_count digit_count digit_count

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

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

CDN - cloudflare - -

ID 格式 msg_01PDNJfQ1gHbVsrFFPa5vE8R msg_018Rxpt2eiXsmnXfeGNedkq8 msg_01BVTBoafFXzPfAzKwoX7Ngc msg_019bNYd4BTwvkyPoXZD4hR3i

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	via: 1.1 Caddy server: uvicorn, 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	d6350724d5ebcc53842094b2644dfe1e
R-ultra	cf-ray: a0d65c1c8ab3fd4f-NRT server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 202606180100553094091128268d9d6k5sLo7yH	id_format: msg_prefix 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_fieldsupstream:anthropic-direct-from-body	a0d65c1c8ab3fd4f-NRT
cc	server: cdn x-new-api-version: v1.0.0-rc.11 x-oneapi-request-id: 20260618010209601246968268d9d6DnWWLkOZ	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_011Cc9pnebuk43tGHgMQmPLU

协议指纹评分（机器自动）

总分 / 100 85 68 76 100

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 8.9 /10 ×20 0.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20

缓存 TTL 一致 0.0 /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

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 5.3 /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

系统提示纯净 5.0 /10 ×8 5.0 /10 ×8 0.0 /10 ×8 10.0 /10 ×8

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

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

身份(自由) 10.0 /10 ×7 0.0 /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

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

流式投递未计入未计入未计入未计入

综合结论

与基线相似度 85 v3.22.0 68 v3.22.0 76 v3.22.0 基准（参考）

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 Thinking 存在 thinking_present
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 8490 ms 10468 ms 6395 ms 5487 ms

输出速度 72.3 tok/s 103.2 tok/s 108.4 tok/s 75.1 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 ⓘ

$0.2034

$0.3378

$0.1245

$0.2002

缓存读取占比 ⓘ 54% 23% 72% 70%

揭晓

通道

服务商 0-0 ↗

通道O-Max

服务商 FoxCode ↗

通道R-ultra

服务商 DawAPI ↗

通道cc

官方基准

返回列表