模型对比 | RPDiag

测试说明点击展开 5 步 quick-probe-v1 详解

同一会话连续 5 步，覆盖 5 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。步与步之间随机延迟 1–4 分钟。第 3 步（cutoff）累计跨度 > 5 分钟保住 sliding cache 检测；全流程总跨度 > 6 分钟容纳新增的第 5 步。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 · 5 步骤

2026/06/16 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

ClaudeZZ

R-kiro

SSSAiCode

R-Kiro

Novaris

O-Max

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 00:37:11 ~ 00:37:21 (6.3s) 00:37:09 ~ 00:37:14 (3.4s) 00:37:15 ~ 00:37:20 (4.6s) 00:37:17 ~ 00:37:27 (5.1s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 4,777 ms 2,298 ms 3,568 ms 3,944 ms

首文本(SSE delta) 4,779 ms 2,417 ms 3,570 ms 3,945 ms

总耗时 5,189 ms 2,421 ms 3,775 ms 4,035 ms

输入 3 0 3 3

输出 5 2 5 5

缓存创建

21,985

5m: 0 / 1h: 21,985

32,318

5m: 0 / 1h: 32,318

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 00:39:53 ~ 00:40:03 (6.5s) 00:38:48 ~ 00:39:33 (40.9s) 00:39:11 ~ 00:39:16 (4.6s) 00:41:13 ~ 00:41:23 (6.0s)

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) 5,167 ms 39,578 ms 3,334 ms 4,539 ms

首文本(SSE delta) 5,218 ms 39,592 ms 3,335 ms 4,541 ms

总耗时 5,576 ms 39,872 ms 3,834 ms 5,035 ms

输入 3 0 3 3

输出 22 18 22 22

缓存创建

196

5m: 0 / 1h: 196

27,015

5m: 27,015 / 1h: 0

189

5m: 0 / 1h: 189

缓存读取 21,985 0 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-4-5

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

步骤 3: cutoff text

时间 00:42:57 ~ 00:43:12 (11.6s) 00:42:07 ~ 00:42:12 (3.5s) 00:42:19 ~ 00:42:24 (4.4s) 00:43:28 ~ 00:43:38 (5.6s)

提示词

查看本步 Prompt

请直接回答你的知识截止日期，格式 YYYY-MM，不要附加任何说明。

HTTP 头(TTFB) 9,462 ms 2,309 ms 3,400 ms 4,079 ms

首文本(SSE delta) 9,465 ms 2,609 ms 3,537 ms 4,358 ms

总耗时 10,479 ms 2,612 ms 3,661 ms 4,609 ms

输入 3 0 3 3

输出 8 3 8 8

缓存创建

5m: 0 / 1h: 81

5m: 44 / 1h: 0

5m: 0 / 1h: 44

缓存读取 22,181 0 27,015 32,507

响应

2025-08

2025-08

2025-01

2025-08

步骤 4: identity_free text

时间 00:45:09 ~ 00:45:19 (8.0s) 00:45:59 ~ 00:46:04 (4.3s) 00:45:39 ~ 00:45:49 (5.7s) 00:46:08 ~ 00:46:18 (5.7s)

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) 5,988 ms 2,015 ms 4,200 ms 3,855 ms

首文本(SSE delta) 6,790 ms 2,219 ms 4,201 ms 3,856 ms

总耗时 7,106 ms 3,524 ms 4,917 ms 4,718 ms

输入 3 0 3 3

输出 66 46 58 58

缓存创建

5m: 0 / 1h: 45

5m: 35 / 1h: 0

5m: 0 / 1h: 35

缓存读取 22,262 0 27,059 32,551

输出速度 - 35.2 tok/s 81.0 tok/s 67.3 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，旨在帮助用户完成各种任务，包括写作、分析、编程和问答等。

我是 Kiro，一款由 Amazon 打造的 AI 驱动开发环境，基于 Anthropic 的 Claude 模型（claude-sonnet-4-6）提供支持，专注于帮助开发者更高效地完成软件工程任务。

我是 Claude，由 Anthropic 开发的 AI 助手。我被设计用来帮助用户完成各种任务，包括写作、分析、编程和回答问题。

我是 Claude，由 Anthropic 开发的 AI 助手。当前版本为 claude-sonnet-4-6，致力于提供安全、有益的对话和任务协助。

步骤 5: knowledge_recall text

时间 00:46:42 ~ 00:47:17 (30.2s) 00:49:22 ~ 00:49:27 (4.4s) 00:49:34 ~ 00:49:49 (10.5s) 00:48:20 ~ 00:48:30 (5.4s)

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) 5,117 ms 1,624 ms 3,301 ms 2,601 ms

首文本(SSE delta) 5,119 ms 1,704 ms 3,679 ms 2,602 ms

总耗时 29,372 ms 3,678 ms 9,793 ms 4,431 ms

输入 3 0 3 3

输出 2,293 221 591 204

缓存创建

6,132

5m: 0 / 1h: 6,132

2,674

5m: 2,674 / 1h: 0

8,033

5m: 0 / 1h: 8,033

缓存读取 2,136 0 0 0

输出速度 94.5 tok/s 112.0 tok/s 96.7 tok/s 111.5 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

总计

总输入(含缓存) 97,018 0 116,175 138,010

总输出 2,394 290 684 297

总缓存创建

28,439

5m: 0 / 1h: 28,439

62,086

5m: 29,768 / 1h: 32,318

40,619

5m: 0 / 1h: 40,619

总缓存读取 68,564 0 54,074 97,376

输出速度 94.5 tok/s 81.4 tok/s 95.0 tok/s 97.4 tok/s

总执行时间 1m 2.5s 56.5s 29.8s 27.8s

总墙钟时间

10m 5.8s

含步骤间等待

12m 17.9s

含步骤间等待

12m 33.8s

含步骤间等待

11m 12.8s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1632

Anthropic: Claude Sonnet 4.6

>=$0.0043

Anthropic: Claude Sonnet 4.6 (成本不完整)

$0.2175

Anthropic: Claude Sonnet 4.6

$0.1860

Anthropic: Claude Sonnet 4.6

通道指纹

来源步骤 knowledge_recall knowledge_recall knowledge_recall knowledge_recall

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

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

CDN - - - -

ID 格式 msg_017QQQtnSTSZhyvfZSQtZFxD msg_84f835d411aa4508b7f99626de26d471 msg_01F1EdCr7UV5heqF249fhjeW msg_01KYh1x2PGMRfDNc2cnZv3vw

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
R-kiro	via: 1.1 google cf-ray: cfd1d3f4c600ecfb4913b6cbfbd37b74-IAD server: nginx	id_format: msg_prefix via_chain: ["1.1 google"] extra_fields: ["diagnostics","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	req_cfd1d3f4c600ecfb4913b6cbfbd37b74 cfd1d3f4c600ecfb4913b6cbfbd37b74-IAD
R-Kiro	server: Photon-Edge x-amzn-requestid: 7a9dc7d6-cb9e-4920-a63f-edd0a7588755	id_format: msg_prefix has_service_tier: false has_inference_geo: false model_has_date_suffix: false usage_has_cache_fields: true	body:id:msg_prefixbody:usage.cache_fieldsupstream:x-amzn-requestidupstream:anthropic-direct-from-body	1781542163187-cxeey6tk31ptp 7a9dc7d6-cb9e-4920-a63f-edd0a7588755
O-Max	server: nginx x-oneapi-request-id: 202606151649349963012598268d9d6TkikAS6R	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	—
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_011Cc5PR5Y4Lfz4r2AyoWrfS

协议指纹评分（机器自动）

总分 / 100 92 46 70 100

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 10.0 /10 ×20 0.0 /10 ×20 5.3 /10 ×20 10.0 /10 ×20

缓存 TTL 一致 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

缓存连续性 10.0 /10 ×14 5.0 /10 ×14 6.7 /10 ×14 10.0 /10 ×14

sliding 5m 10.0 /10 ×13 5.0 /10 ×13 7.7 /10 ×13 10.0 /10 ×13

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

系统提示纯净 0.0 /10 ×8 未计入 10.0 /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 0.0 /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 0.0 /10 ×6 10.0 /10 ×6 10.0 /10 ×6

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

综合结论

与基线相似度 92 v3.16.0 46 v3.16.0 70 v3.16.0 基准（参考）

雷达图例

共 19 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 中文遵循 instruction_following_lang
19 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 11544 ms 10421 ms 5196 ms 4566 ms

输出速度 94.5 tok/s 81.4 tok/s 95.0 tok/s 97.4 tok/s

完成步骤 5/5 成功 5/5 成功 5/5 成功 5/5 成功

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.1632

>=$0.0043 (成本不完整)

$0.2175

$0.1860

缓存读取占比 ⓘ 71% - 47% 71%

揭晓

通道

服务商 ClaudeZZ ↗

通道R-kiro

服务商 SSSAiCode ↗

通道R-Kiro

服务商 Novaris ↗

通道O-Max

官方基准

返回列表