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

2026/06/20 运行日期

claude-opus-4-8 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

SSSAiCode

O-Max-L2

LingxiCode

M-

HongMaCC

u-cc

官方基准

SSSAiCode

R-Kiro

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 17:28:55 ~ 17:29:05 (9.0s) 17:28:49 ~ 17:28:54 (4.7s) 17:28:51 ~ 17:28:56 (4.5s) 17:28:57 ~ 17:29:02 (3.3s) 17:28:53 ~ 17:29:03 (6.0s)

提示词

查看本步 Prompt

ping. Reply with the single word: pong

HTTP 头(TTFB) 7,794 ms 3,228 ms 2,864 ms 2,094 ms 2,544 ms

首文本(SSE delta) 7,944 ms 3,235 ms 2,866 ms 2,096 ms 2,575 ms

总耗时 8,076 ms 3,438 ms 2,866 ms 2,255 ms 2,692 ms

输入 1,428 1,454 2 1,454 0

输出 4 4 4 4 4

缓存创建

29,765

5m: 0 / 1h: 29,765

30,223

5m: 0 / 1h: 30,223

31,079

5m: 0 / 1h: 31,079

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 17:31:16 ~ 17:31:21 (3.5s) 17:31:35 ~ 17:31:45 (5.2s) 17:31:31 ~ 17:31:36 (4.2s) 17:30:32 ~ 17:30:37 (3.6s) 17:31:09 ~ 17:31:14 (3.4s)

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) 2,144 ms 3,565 ms 2,990 ms 2,344 ms 2,473 ms

首文本(SSE delta) 2,518 ms 3,566 ms 2,992 ms 2,345 ms 2,475 ms

总耗时 2,583 ms 3,977 ms 2,992 ms 2,633 ms 2,616 ms

输入 2 2 2 2 0

输出 30 28 28 28 28

缓存创建

31,376

5m: 0 / 1h: 31,376

1,646

5m: 0 / 1h: 1,646

194

5m: 0 / 1h: 194

1,646

5m: 0 / 1h: 1,646

缓存读取 0 30,223 31,079 30,222 0

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 17:35:15 ~ 17:35:20 (4.1s) 17:33:01 ~ 17:33:06 (4.7s) 17:35:22 ~ 17:35:27 (4.3s) 17:32:07 ~ 17:32:12 (3.2s) 17:32:58 ~ 17:33:03 (3.3s)

提示词

查看本步 Prompt

Knowledge cutoff date? Reply with YYYY-MM only.

HTTP 头(TTFB) 3,154 ms 3,520 ms 3,206 ms 1,876 ms 2,351 ms

首文本(SSE delta) 3,292 ms 3,521 ms 3,207 ms 1,878 ms 2,413 ms

总耗时 3,385 ms 3,798 ms 3,207 ms 2,141 ms 2,496 ms

输入 2 2 2 2 0

输出 6 6 6 6 8

缓存创建

5m: 0 / 1h: 56

5m: 0 / 1h: 64

5m: 0 / 1h: 62

5m: 0 / 1h: 54

缓存读取 31,376 31,869 31,273 31,868 0

响应

2025-03

2026-01

2026-01

2026-01

2026-01

步骤 4: identity_free text

时间 17:37:49 ~ 17:37:54 (4.6s) 17:36:08 ~ 17:36:18 (5.1s) 17:39:05 ~ 17:39:15 (4.9s) 17:34:03 ~ 17:34:08 (4.0s) 17:34:36 ~ 17:34:41 (3.9s)

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) 2,968 ms 3,474 ms 3,781 ms 1,960 ms 2,524 ms

首文本(SSE delta) 3,811 ms 3,475 ms 3,783 ms 1,962 ms 2,530 ms

总耗时 3,880 ms 4,282 ms 3,783 ms 2,915 ms 2,980 ms

输入 2 2 2 2 0

输出 49 48 45 57 49

缓存创建

5m: 0 / 1h: 35

5m: 0 / 1h: 39

5m: 0 / 1h: 37

5m: 0 / 1h: 35

缓存读取 31,432 31,933 31,335 31,922 0

输出速度 - 59.5 tok/s burst 59.8 tok/s 108.9 tok/s

响应

我是 Claude，由 Anthropic 公司打造的 AI 助手，专注于帮助用户完成软件工程等各类任务。

我是 Claude，由 Anthropic 开发的 AI 助手，可帮助你完成编程、写作、分析等各类任务。

我是 Claude（Opus 4.8 模型），属于 Claude 产品系列，由 Anthropic 公司开发。

我是 Claude，由 Anthropic 公司打造的 AI 助手，当前模型为 Claude Opus 4.8，可协助编程与各类任务。

我是 Claude，由 Anthropic 开发的 AI 助手。我可以帮你写代码、分析问题、研究资料和处理各类软件工程任务。

步骤 5: knowledge_recall text

时间 17:40:52 ~ 17:41:02 (5.1s) 17:38:08 ~ 17:38:23 (10.1s) 17:40:29 ~ 17:40:44 (10.8s) 17:36:58 ~ 17:37:08 (8.8s) 17:38:18 ~ 17:38:33 (10.9s)

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) 1,988 ms 3,341 ms 9,536 ms 1,847 ms 6,169 ms

首文本(SSE delta) 1,995 ms 6,535 ms 9,537 ms 5,582 ms 7,424 ms

总耗时 4,287 ms 9,189 ms 9,537 ms 7,820 ms 9,867 ms

输入 2 2 2 2 0

输出 341 708 745 740 342

缓存创建

4,270

5m: 0 / 1h: 4,270

4,774

5m: 0 / 1h: 4,774

3,713

5m: 0 / 1h: 3,713

4,768

5m: 0 / 1h: 4,768

缓存读取 0 0 15,933 0 0

输出速度 148.8 tok/s burst burst burst 140.0 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

{"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 17:43:09 ~ 17:43:14 (1.0s) 17:40:27 ~ 17:40:32 (3.2s) 17:42:28 ~ 17:42:43 (13.8s) 17:40:54 ~ 17:40:59 (2.7s) 17:40:48 ~ 17:41:03 (10.7s)

提示词

查看本步 Prompt

Count digits exactly in s. Return {"n":len(s),"c":[count0,count7,count9]}. s=316508295840468820967713783309271751724277493651897477750926116224533702

HTTP 头(TTFB) - 1,910 ms 12,441 ms 1,111 ms 4,740 ms

首文本(SSE delta) - 1,911 ms 12,442 ms 1,112 ms 9,736 ms

总耗时 - 2,359 ms 12,442 ms 1,531 ms 9,881 ms

输入 - 211 2 211 0

输出 - 19 1,067 19 20

缓存创建

5m: 0 / 1h: 0

18,544

5m: 0 / 1h: 18,544

5m: 0 / 1h: 0

缓存读取 - 0 0 0 0

输出速度 - - burst - -

响应

失败

评测失败

{"n":70,"c":[3,9,7]}

{"n":72,"c":[6,13,6]}

{"n":70,"c":[3,12,7]}

{"n":72,"c":[6,13,6]}

总计

总输入(含缓存)

129,746

仅前 6 步

132,444

163,261

132,410

总输出

430

仅前 6 步

813

1,895

854

451

总缓存创建

65,502

5m: 0 / 1h: 65,502

仅前 6 步

36,746

5m: 0 / 1h: 36,746

53,629

5m: 0 / 1h: 53,629

36,725

5m: 0 / 1h: 36,725

总缓存读取

62,808

仅前 6 步

94,025

109,620

94,012

输出速度

148.8 tok/s

仅前 6 步

59.5 tok/s

burst

59.8 tok/s

135.2 tok/s

总执行时间

27.3s

仅前 6 步

33.0s

42.5s

25.7s

38.2s

总墙钟时间

14m 18.9s

含步骤间等待

仅前 6 步

11m 43.0s

含步骤间等待

13m 51.2s

含步骤间等待

12m 1.9s

含步骤间等待

12m 10.0s

含步骤间等待

按官方价目重估

同等内容若直发 Anthropic 的估算 ⓘ

$0.4587

Anthropic: Claude Opus 4.8 (部分)

$0.3054

Anthropic: Claude Opus 4.8

$0.4374

Anthropic: Claude Opus 4.8

$0.3063

Anthropic: Claude Opus 4.8

>=$0.0113

Anthropic: Claude Opus 4.8 (成本不完整)

通道指纹

来源步骤 knowledge_recall digit_count digit_count digit_count digit_count

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

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

CDN - cloudflare - - -

ID 格式 msg_01Fck2gbNdYWwTSrzV8Btam1 msg_01Vy3VontXUotevne7KPSv7u msg_014E6nb4FQEhMqdxtrQe6Uwj msg_01XMw7k9Zo2gWcHYkSps7NK6 msg_66b317d887124f38a2f705fb8d615063

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: a0e9d08c8d244adb-LAX server: Photon-Edge set-cookie: _cfuvid=odtXMeac0pLCSfGLF1WGbetZ39fT.D5cZLLZOCqFC3s-1781948453.8429527-1.0.1.1-xdZvP6WbtefiHo5BBit08gHQs_6xr9DVuHZrcfKCSw4; 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	1781948453801-d8jt6tu62sqz4 req_011CcEHs3M9xyXWje3Jmn9Em a0e9d08c8d244adb-LAX
M-	cf-ray: a0e9cfef18f1fd4f-NRT server: cloudflare set-cookie: server_name_session=e44607f7628a2aebd93b10e6388b22e1; Max-Age=86400; httponly; path=/ x-oneapi-request-id: 20260620094029280431925LJrk3Ak0	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-idhdr:server:cloudflarehdr:cf-raybody:id:msg_prefixbody:usage.service_tierbody:usage.inference_geobody:usage.cache_fieldsbody:extra_fieldsupstream:anthropic-direct-from-body	a0e9cfef18f1fd4f-NRT
u-cc	server: nginx x-new-api-version: 1.0.14 x-oneapi-request-id: 202606200942294548789088268d9d6thp6kliL	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_011CcEHsChUgtMnoFyJwcWwg
R-Kiro	server: Photon-Edge x-amzn-requestid: 5d067fe3-a1e3-4a40-96c5-044d3f2e2961	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	1781948449817-w67h4azk8dxb6 5d067fe3-a1e3-4a40-96c5-044d3f2e2961

协议指纹评分（机器自动）

总分 / 100 98 88 100 57

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 5.0 /10 ×14

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

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

缓存命中比 — 10.0 /10 ×20 10.0 /10 ×20 10.0 /10 ×20 0.0 /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 10.0 /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 — 10.0 /10 ×8 10.0 /10 ×8 10.0 /10 ×8 0.0 /10 ×8

Req-ID 透传 — 0.0 /10 ×4 0.0 /10 ×4 10.0 /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 10.0 /10 ×6 0.0 /10 ×6

inference_geo — 10.0 /10 ×5 10.0 /10 ×5 10.0 /10 ×5 0.0 /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 5.0 /10 ×8 10.0 /10 ×8 未计入

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 0.0 /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 98 v3.24.0 88 v3.24.0 基准（参考） 57 v3.24.0

雷达图例

共 22 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 系统提示纯净 system_prompt_clean
9 Service Tier service_tier_present
10 缓存命中比 cache_hit_ratio_match
11 缓存 TTL 一致 cache_ttl_consistency
12 推理地区 inference_geo_present
13 缓存连续 cache_continuity_intra
14 知识召回 knowledge_recall_match
15 延迟基线 latency_baseline_match
16 原生 msg-ID anthropic_msg_id_format
17 Sliding 缓存 cache_sliding_correctness
18 结构化身份 identity_structured_match
19 信封自报一致 envelope_self_report_match
20 中文遵循 instruction_following_lang
21 档位思考量 tier_thinking_volume_match
22 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 4442 ms 4507 ms 5805 ms 3216 ms 5089 ms

输出速度 148.8 tok/s 59.5 tok/s burst 59.8 tok/s 135.2 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 ⓘ

$0.4587 (部分)

$0.3054

$0.4374

$0.3063

>=$0.0113 (成本不完整)

缓存读取占比 ⓘ 48% (部分) 71% 67% 71% -

揭晓

通道

服务商 SSSAiCode ↗

通道O-Max-L2

服务商 LingxiCode ↗

通道M-

服务商 HongMaCC ↗

通道u-cc

官方基准

服务商 SSSAiCode ↗

通道R-Kiro

返回列表