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

2026/06/29 运行日期

claude-opus-4-8 官方基准

2 通道分数 < 70 另含 1 任务失败

已揭晓揭晓进度

步骤 / 指标

SSSAiCode

O-Max

SSSAiCode

R-Kiro

ClaudeZZ

R-kiro

FoxCode

官方基准

基本信息

服务名 Claude Code Claude Code Claude Code Claude Code Claude Code

模型 claude-opus-4-8 claude-opus-4-8 claude-opus-4-8 claude-opus-4-8 claude-opus-4-8

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 11:20:35 ~ 11:20:40 (3.0s) 11:20:37 ~ 11:20:42 (3.0s) 11:20:31 ~ 11:20:36 (4.7s) 11:20:33 ~ 11:20:38 (3.1s) 11:20:39 ~ 11:20:44 (3.3s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 1,928 ms 2,115 ms 3,192 ms 1,858 ms 2,102 ms

首文本(SSE delta) 2,022 ms 2,132 ms 3,203 ms 1,859 ms 2,104 ms

总耗时 2,067 ms 2,135 ms 3,618 ms 2,205 ms 2,315 ms

输入 1,428 0 2 1,454 1,454

输出 4 4 4 4 4

缓存创建

29,763

5m: 0 / 1h: 29,763

27,913

5m: 0 / 1h: 27,913

30,222

5m: 30,222 / 1h: 0

30,222

5m: 0 / 1h: 30,222

缓存读取 0 0 0 0 0

响应

pong

pong

pong

pong

pong

步骤 2: identity text

时间 11:24:23 ~ 11:24:28 (3.2s) 11:23:37 ~ 11:23:42 (3.0s) 11:21:59 ~ 11:22:04 (4.4s) 11:21:39 ~ 11:21:49 (5.0s) 11:21:45 ~ 11:21:55 (6.0s)

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。请按下列 3 行格式精确填空作答（label 不变，值小写）：
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) 1,951 ms 2,037 ms 3,459 ms 3,498 ms 4,595 ms

首文本(SSE delta) 2,261 ms 2,052 ms 3,460 ms 3,499 ms 4,596 ms

总耗时 2,281 ms 2,172 ms 3,672 ms 4,033 ms 4,975 ms

输入 2 0 2 2 2

输出 30 26 28 28 28

缓存创建

1,608

5m: 0 / 1h: 1,608

194

5m: 0 / 1h: 194

1,634

5m: 1,634 / 1h: 0

1,634

5m: 0 / 1h: 1,634

缓存读取 29,763 0 27,913 30,222 30,222

输出速度 burst - - - -

响应

vendor: anthropic
brand: claude
model: claude-opus-4-20250514

vendor: anthropic
brand: claude
model: claude-opus-4-8

vendor: anthropic
brand: claude
model: claude-opus-4-8

vendor: anthropic
brand: claude
model: claude-opus-4-8

vendor: anthropic
brand: claude
model: claude-opus-4-8

步骤 3: cutoff text

时间 11:27:59 ~ 11:28:04 (3.1s) 11:26:43 ~ 11:26:48 (2.7s) 11:25:16 ~ 11:25:26 (6.4s) 11:23:28 ~ 11:23:33 (2.9s) 11:24:41 ~ 11:24:46 (3.0s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 1,865 ms 1,808 ms 4,507 ms 1,741 ms 1,767 ms

首文本(SSE delta) 2,032 ms 1,832 ms 5,201 ms 1,742 ms 1,769 ms

总耗时 2,188 ms 1,963 ms 5,467 ms 2,141 ms 2,020 ms

输入 2 0 2 2 2

输出 6 6 124 6 6

缓存创建

5m: 0 / 1h: 56

5m: 0 / 1h: 98

5m: 62 / 1h: 0

5m: 0 / 1h: 54

缓存读取 31,371 0 28,107 31,856 31,856

输出速度 - - burst - -

响应

2025-01

2026-01

2025-03

2026-01

2026-01

步骤 4: identity_free text

时间 11:30:14 ~ 11:30:24 (7.6s) 11:28:34 ~ 11:28:39 (3.9s) 11:27:44 ~ 11:27:55 (7.0s) 11:25:37 ~ 11:25:42 (4.1s) 11:27:07 ~ 11:27:12 (3.8s)

提示词

查看本步 Prompt

请用中文介绍你自己是什么产品或助手，由谁开发，50 字以内。

HTTP 头(TTFB) 5,563 ms 2,250 ms 4,811 ms 2,162 ms 1,480 ms

首文本(SSE delta) 6,647 ms 3,009 ms 4,812 ms 2,164 ms 1,482 ms

总耗时 6,792 ms 3,144 ms 6,099 ms 3,257 ms 2,704 ms

输入 2 0 2 2 2

输出 68 60 77 61 75

缓存创建

5m: 0 / 1h: 39

172

5m: 0 / 1h: 172

5m: 31 / 1h: 0

5m: 0 / 1h: 31

缓存读取 31,427 0 28,205 31,918 31,910

输出速度 - - 59.8 tok/s 55.8 tok/s 61.4 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手。我可以协助编程、写作、分析和回答各类问题，致力于提供有帮助、诚实且安全的服务。

我是 Claude，由 Anthropic 开发的 AI 助手。我可以帮你处理编程、写作、分析、研究等各类任务，擅长理解需求并协作完成软件工程工作。

我是 Claude，由 Anthropic 开发的 AI 助手。当前以 Claude Code 命令行工具的形式运行，专注于协助软件工程任务，比如编写代码、调试和解答技术问题。

我是 Claude，一个由 Anthropic 开发的 AI 助手，基于 Claude Opus 4.8 模型，可协助完成软件工程及各类任务。

我是 Claude，由 Anthropic 公司开发的人工智能助手，具体模型为 Opus 4.8（claude-opus-4-8）。我可以帮你处理软件工程、编程及各类任务。

步骤 5: knowledge_recall text

时间 11:33:12 ~ 11:33:22 (6.4s) 11:31:46 ~ 11:32:01 (10.5s) 11:31:18 ~ 11:31:28 (9.2s) 11:28:43 ~ 11:28:58 (10.1s) 11:30:55 ~ 11:31:05 (8.6s)

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) 2,640 ms 6,507 ms 3,626 ms 2,910 ms 1,455 ms

首文本(SSE delta) 5,389 ms 9,108 ms 5,818 ms 6,882 ms 5,211 ms

总耗时 5,612 ms 9,712 ms 8,423 ms 9,198 ms 7,602 ms

输入 2 0 2 2 2

输出 317 342 633 686 706

缓存创建

4,288

5m: 0 / 1h: 4,288

11,278

5m: 0 / 1h: 11,278

4,764

5m: 4,764 / 1h: 0

4,770

5m: 0 / 1h: 4,770

缓存读取 0 0 0 0 0

输出速度 - burst burst burst burst

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

{"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 11:35:52 ~ 11:35:57 (1.1s) 11:33:17 ~ 11:33:32 (13.7s) 11:33:07 ~ 11:33:27 (17.9s) 11:31:42 ~ 11:31:47 (2.9s) 11:33:35 ~ 11:33:40 (2.4s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=325066596191753585370738579884214324148541333028577138004860501619173803

HTTP 头(TTFB) - 4,625 ms 3,692 ms 1,616 ms 961 ms

首文本(SSE delta) - 12,759 ms 16,932 ms 1,618 ms 962 ms

总耗时 - 12,906 ms 17,139 ms 2,006 ms 1,417 ms

输入 - 0 2 211 211

输出 - 20 1,504 19 19

缓存创建

5,091

5m: 0 / 1h: 5,091

5m: 0 / 1h: 0

缓存读取 - 0 3,067 0 0

输出速度 - - burst - -

响应

失败

评测失败

{"n":72,"c":[8,7,4]}

{"n":72,"c":[8,7,4]}

{"n":70,"c":[5,5,4]}

{"n":70,"c":[6,5,3]}

总计

总输入(含缓存)

129,751

仅前 6 步

132,050

132,382

132,372

总输出

425

仅前 6 步

458

2,370

804

838

总缓存创建

35,754

5m: 0 / 1h: 35,754

仅前 6 步

44,746

5m: 0 / 1h: 44,746

36,713

5m: 36,713 / 1h: 0

36,711

5m: 0 / 1h: 36,711

总缓存读取

92,561

仅前 6 步

87,292

93,996

93,988

输出速度

burst

仅前 6 步

burst

59.8 tok/s

55.8 tok/s

61.4 tok/s

总执行时间

24.4s

仅前 6 步

36.9s

49.6s

28.1s

27.2s

总墙钟时间

15m 21.9s

含步骤间等待

仅前 6 步

12m 54.9s

含步骤间等待

12m 56.1s

含步骤间等待

11m 13.9s

含步骤间等待

13m 1.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2875

Anthropic: Claude Opus 4.8 (部分)

>=$0.0115

Anthropic: Claude Opus 4.8 (成本不完整)

$0.3826

Anthropic: Claude Opus 4.8

$0.2189

Anthropic: Claude Opus 4.8

$0.3058

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 knowledge_recall digit_count digit_count digit_count digit_count

平台 direct (100%) direct (75%) direct (100%) direct (100%) direct (100%)

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

CDN - - - cloudflare -

ID 格式 msg_01RucZUg5kLber2gGR4PQ14J msg_e1b02f54c722405895964554a43c1346 msg_0164a35zpuXUGvLBEnYPz59n msg_01ENLcx5JFNxDrVvrkyPhYno msg_01Da628BQ6tc6NAuYJ2iB33R

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: a131de55199a75b7-LAX server: Photon-Edge set-cookie: _cfuvid=Gz8_jcJmT7H726UoF7NwghqG4yPMFo1vIyAT7Le38no-1782703993.1333728-1.0.1.1-F0VWbLlSZ1HpM4E0ILYrNOf49RESXcT75KQ41OTDvkE; 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	body:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	1782703993033-woz26awrrovrr req_011CcWr76RNsUhRvFu6Jtbir a131de55199a75b7-LAX
R-Kiro	server: Photon-Edge x-amzn-requestid: 40da6e51-555d-49ec-97a9-af53549015da	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	1782703998132-0mocgrqte8od6 40da6e51-555d-49ec-97a9-af53549015da
R-kiro	via: 1.1 google cf-ray: 74a4439d68fc20ae6ed570705eadd360-IAD server: nginx	id_format: msg_prefix via_chain: ["1.1 google"] 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	req_74a4439d68fc20ae6ed570705eadd360 74a4439d68fc20ae6ed570705eadd360-IAD
cc	cf-ray: a131dc22daafe3bf-NRT server: cloudflare set-cookie: _cfuvid=j.2lHsFLAZKeXjkgMZjHR0yHcaR6hYYPb.CYuRMGewM-1782703903.4944046-1.0.1.1-8D8Gf16E3PIz1BaMj9nQermhiRIBxi0S6eWdqwhdR90; 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_011CcWqzUuXwEPvdAyk8Sfde a131dc22daafe3bf-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_011CcWr8pLYzfYyZJtxXxsJv

协议指纹评分（机器自动）

总分 / 100 50 80 91 100

基线状态已对照基线已对照基线已对照基线已对照基线

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 — 5.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14 10.0 /10 ×14

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

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

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

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 0.0 /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

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

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

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

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

inference_geo — 0.0 /10 ×5 10.0 /10 ×5 10.0 /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 — 未计入 0.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8

延迟基线 — 10.0 /10 ×5 6.9 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

流式投递 — 0.1 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5

综合结论

与基线相似度执行失败，不评分 50 v3.25.0 80 v3.25.0 91 v3.25.0 基准（参考）

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 Thinking 存在 thinking_present
5 流式投递 buffer_dump_match
6 自由身份 identity_free_clean
7 停止原因 stop_reason_present
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 3788 ms 5339 ms 7403 ms 3807 ms 3506 ms

输出速度 burst burst 59.8 tok/s 55.8 tok/s 61.4 tok/s

完成步骤 5/6 成功 · 1 失败 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 ⓘ

$0.2875 (部分)

>=$0.0115 (成本不完整)

$0.3826

$0.2189

$0.3058

缓存读取占比 ⓘ 71% (部分) - 66% 71% 71%

揭晓

通道

服务商 SSSAiCode ↗

通道O-Max

服务商 SSSAiCode ↗

通道R-Kiro

服务商 ClaudeZZ ↗

通道R-kiro

服务商 FoxCode ↗

通道cc

官方基准

返回列表