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

2026/06/25 运行日期

claude-opus-4-8 官方基准

2 通道分数 < 70 另含 2 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

TopRouterCN / O-Max
```
评测失败
```

步骤 / 指标

SSSAiCode

O-Max-L2

SSSAiCode

R-Kiro

AIMZ

M-Max

官方基准

基本信息

服务名 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

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 10:58:45 ~ 10:58:50 (3.7s) 10:58:49 ~ 10:59:09 (18.3s) 10:58:47 ~ 10:58:52 (4.6s) 10:58:51 ~ 10:58:56 (4.1s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 2,429 ms 11,665 ms 2,699 ms 2,676 ms

首文本(SSE delta) 2,523 ms 17,070 ms 2,728 ms 2,678 ms

总耗时 2,654 ms 17,315 ms 2,842 ms 2,814 ms

输入 1,428 0 1,428 1,454

输出 4 4 4 4

缓存创建

29,762

5m: 0 / 1h: 29,762

29,758

5m: 0 / 1h: 29,758

30,222

5m: 0 / 1h: 30,222

缓存读取 0 0 0 0

响应

pong

pong

pong

pong

步骤 2: identity text

时间 11:02:29 ~ 11:02:34 (4.2s) 11:00:34 ~ 11:00:39 (3.4s) 11:02:00 ~ 11:02:05 (4.6s) 11:01:22 ~ 11:01:27 (3.7s)

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) 2,633 ms 2,382 ms 2,596 ms 2,317 ms

首文本(SSE delta) 2,990 ms 2,404 ms 3,027 ms 2,318 ms

总耗时 3,478 ms 2,620 ms 3,101 ms 2,694 ms

输入 2 0 2 2

输出 30 28 32 28

缓存创建

1,620

5m: 0 / 1h: 1,620

1,620

5m: 0 / 1h: 1,620

1,646

5m: 0 / 1h: 1,646

缓存读取 29,762 0 29,758 30,222

输出速度 61.5 tok/s - - -

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-4-1-20250805

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

步骤 3: cutoff text

时间 11:05:42 ~ 11:05:47 (3.6s) 11:04:36 ~ 11:04:41 (3.1s) 11:05:17 ~ 11:05:22 (4.7s) 11:02:52 ~ 11:02:57 (3.5s)

提示词

查看本步 Prompt

你掌握的世界知识截止于何时？YYYY-MM 格式作答，不解释。

HTTP 头(TTFB) 2,712 ms 2,163 ms 2,272 ms 2,336 ms

首文本(SSE delta) 2,818 ms 2,185 ms 2,428 ms 2,340 ms

总耗时 2,935 ms 2,251 ms 2,506 ms 2,560 ms

输入 2 0 2 2

输出 6 8 6 6

缓存创建

5m: 0 / 1h: 64

5m: 0 / 1h: 66

5m: 0 / 1h: 62

缓存读取 31,382 0 31,378 31,868

响应

2025-01

2026-01

2025-01

2025-01

步骤 4: identity_free text

时间 11:09:20 ~ 11:09:30 (7.5s) 11:07:25 ~ 11:07:30 (4.8s) 11:09:17 ~ 11:09:22 (4.8s) 11:04:03 ~ 11:04:13 (9.7s)

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) 5,217 ms 2,851 ms 2,753 ms 7,323 ms

首文本(SSE delta) 6,329 ms 2,879 ms 3,554 ms 7,324 ms

总耗时 6,485 ms 3,947 ms 3,632 ms 8,702 ms

输入 2 0 2 2

输出 69 63 50 81

缓存创建

5m: 0 / 1h: 39

5m: 0 / 1h: 35

5m: 0 / 1h: 39

缓存读取 31,446 0 31,444 31,930

输出速度 - 59.0 tok/s - 58.8 tok/s

响应

我是 Claude，由 Anthropic 开发的 AI 助手。我可以帮助处理编程、写作、分析等各类任务，致力于提供有帮助、诚实且安全的回应。

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，擅长理解需求并给出实用解答。很高兴为你服务。

我是 Claude，由 Anthropic 公司打造的人工智能助手，可协助完成软件工程、写作、分析等多种任务。

我是 Claude，由 Anthropic 开发的 AI 助手，基于 Claude Opus 4.8 模型。我能协助编程、写作、分析等多种任务，致力于提供有用、诚实且安全的帮助。

步骤 5: knowledge_recall text

时间 11:11:23 ~ 11:11:34 (5.7s) 11:09:00 ~ 11:09:15 (10.7s) 11:10:38 ~ 11:10:48 (5.8s) 11:07:00 ~ 11:07:15 (10.2s)

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) 2,228 ms 5,163 ms 1,838 ms 2,198 ms

首文本(SSE delta) 4,735 ms 5,473 ms 4,462 ms 6,881 ms

总耗时 4,897 ms 9,780 ms 4,606 ms 9,005 ms

输入 2 0 2 2

输出 317 342 317 828

缓存创建

4,308

5m: 0 / 1h: 4,308

4,283

5m: 0 / 1h: 4,283

4,804

5m: 0 / 1h: 4,804

缓存读取 0 0 0 0

输出速度 - 79.4 tok/s 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

步骤 6: digit_count text

时间 11:13:57 ~ 11:14:02 (1.1s) 11:13:06 ~ 11:13:17 (9.1s) 11:13:52 ~ 11:13:57 (3.0s) 11:09:13 ~ 11:09:19 (2.8s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=474614645825238025691732054422770662109854419981231658476817534034324093

HTTP 头(TTFB) - 5,196 ms 1,418 ms 1,361 ms

首文本(SSE delta) - 8,157 ms 1,419 ms 1,363 ms

总耗时 - 8,403 ms 1,972 ms 1,776 ms

输入 - 0 140 211

输出 - 20 19 19

缓存创建

5m: 0 / 1h: 0

缓存读取 - 0 0 0

响应

失败

评测失败

{"n":72,"c":[6,6,5]}

{"n":72,"c":[4,5,6]}

{"n":72,"c":[5,5,4]}

总计

总输入(含缓存)

129,819

仅前 6 步

129,918

132,466

总输出

426

仅前 6 步

465

428

966

总缓存创建

35,793

5m: 0 / 1h: 35,793

仅前 6 步

35,762

5m: 0 / 1h: 35,762

36,773

5m: 0 / 1h: 36,773

总缓存读取

92,590

仅前 6 步

92,580

94,020

输出速度

61.5 tok/s

仅前 6 步

75.3 tok/s

burst

58.8 tok/s

总执行时间

25.9s

仅前 6 步

49.4s

27.6s

34.1s

总墙钟时间

15m 17.0s

含步骤间等待

仅前 6 步

14m 27.9s

含步骤间等待

15m 10.0s

含步骤间等待

10m 28.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2878

Anthropic: Claude Opus 4.8 (部分)

>=$0.0116

Anthropic: Claude Opus 4.8 (成本不完整)

$0.2884

Anthropic: Claude Opus 4.8

$0.3094

Anthropic: Claude Opus 4.8

通道指纹

来源步骤 knowledge_recall digit_count digit_count digit_count

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 - - cloudflare -

ID 格式 msg_018PJHWRGcNzKXQcZfbNSthF msg_fc607c9cb1a04c3896c7d39a931e7e90 msg_01LvgTNCnBuy99M9Ad3cHFoM msg_019FqKKQNmqX5ssW8ABQ719o

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-L2	cf-ray: a110c8e81dc12b84-LAX server: Photon-Edge set-cookie: _cfuvid=Ha8vizTKxVINIa4dh5htlZusGPWzrgdQi2n0A8oINW8-1782357085.4587946-1.0.1.1-i.jjsO6Ulyy9toDk97aaKExi43bhIEvPUCNhmnpt4.k; 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	1782357085052-94i400k56mts5 req_011CcPFD3817JXUMho4CbKWk a110c8e81dc12b84-LAX
R-Kiro	server: Photon-Edge x-amzn-requestid: b0d21325-5d14-4648-b070-40d5f0cfcbde	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	1782357187718-c58d56l8ohio4 b0d21325-5d14-4648-b070-40d5f0cfcbde
M-Max	cf-ray: a110cc83adf99918-AMS server: cloudflare x-new-api-version: v0.0.0 x-oneapi-request-id: 202606250313532413433708268d9d6PTbgC6fm	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	a110cc83adf99918-AMS
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_011CcPF3SBJUWxafopHFZwyp

协议指纹评分（机器自动）

总分 / 100 55 80 93

基线状态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

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

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

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

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

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

知识截止 — 10.0 /10 ×7 0.0 /10 ×7 0.0 /10 ×7

身份(自由) — 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

Req-ID 透传 — 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

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

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

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

系统提示纯净 — 未计入 5.0 /10 ×8 10.0 /10 ×8

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

流式投递 — 10.0 /10 ×5 0.0 /10 ×5 10.0 /10 ×5

综合结论

与基线相似度执行失败，不评分 55 v3.24.1 80 v3.24.1 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 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 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 4090 ms 7386 ms 3110 ms 4592 ms

输出速度 61.5 tok/s 75.3 tok/s burst 58.8 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

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.2878 (部分)

>=$0.0116 (成本不完整)

$0.2884

$0.3094

缓存读取占比 ⓘ 71% (部分) - 71% 71%

揭晓

通道

服务商 SSSAiCode ↗

通道O-Max-L2

服务商 SSSAiCode ↗

通道R-Kiro

服务商 AIMZ ↗

通道M-Max

官方基准

返回列表